Artificial Intelligence
Transferability for General Reasoning: An Automated Curriculum for Multi-Domain RLVR
Yongjin Yang, Jiarui Liu, Yinghui He et al. (+3 more)
Reinforcement learning with verifiable rewards (RLVR) has been extended from single-domain training to multi-domain
reasoning suites spanning mathematics, programming, and science. However, the training curriculum (how often each domain is sampled) is typically fixed or hand-tuned, even though
reasoning skills transfer unevenly across domains. Existing learnability-based curricula adapt to where the policy is currently improving, but are blind to whether a gradient step on the selected domain benefits the remaining domains. In this paper, we propose Transfer-Aware Curriculum (TAC), a bandit-style online curriculum that prioritizes domains whose updates broadly benefit the rest of the training suite. TAC repurposes signals already produced by RL training: per-domain advantages capture local learnability, and projected gradients, taken from the GRPO step being computed, estimate cross-domain transferability via gradient-geometry
alignment, at negligible cost (<1% wall-clock overhead). Across a six-domain
reasoning suite, TAC achieves the best macro-averaged accuracy on both Qwen3-1.7B and Llama3.2-3B, outperforming proportional random sampling, a hand-designed schedule, and a learnability-only bandit, and improving over the last of these by up to 2.8 points (10% relative). Ablations show performance degrades sharply when the transferability term is removed, and TAC remains robust on imbalanced training mixtures where learnability-only curricula over-commit to dominant domains. Our findings establish cross-domain transferability as a key signal for curriculum design in multi-domain RLVR.
Read the full abstract ▾
All authors Yongjin Yang, Jiarui Liu, Yinghui He, Lechen Zhang, Bernhard Schölkopf, Zhijing Jin
Artificial Intelligence
Matilda: Engine-Agnostic Search with Human Policy Guidance
Jason Carlson
Chess engines have evolved from search-based systems optimized for strength to neural policies optimized for predicting human decisions. Existing approaches largely separate these goals: search engines achieve superhuman strength but poorly model humans, while models such as Maia-3 capture rating-conditioned behavior yet degrade at elite levels. We present Matilda, a modular residual re-ranking architecture that decouples behavioral priors from tactical search, combining a frozen human policy with an engine-agnostic search backend through a lightweight residual model. Matilda learns residual corrections over the full legal-move distribution from frozen policy context, time control, player-style
embeddings, and search-derived candidate features. A zero-initialized residual head exactly recovers the frozen policy before training while optimization minimizes negative log-likelihood (NLL). Instantiated with Maia-3 and Stockfish, Matilda reduces human-move prediction NLL by 18.5% and raises top-1 accuracy from 60.1% to 66.1% on temporally held-out verified-human 3000+ Elo Lichess blitz games, with player-style
embeddings contributing a further 1.8% and +0.2 percentage points (pp) respectively. Seed-paired ablations attribute these gains to search rather than additional data; the findings are replicated in Go – decomposing expert play into recognition and verified calculation. Below 2500 Elo, where search annotations are unavailable, Matilda preserves Maia-3's performance.
Read the full abstract ▾
Artificial Intelligence
TRUSTMEM: Learning Trustworthy Memory Consolidation for LLM Agents with Long-Term Memory
Tianyu Yang, Sudipta Paul, Vijay Srinivasan et al. (+2 more)
Large
language model (LLM)
agents rely on long-term
memory to support extended interactions and personalized assistance beyond finite context windows. Existing
memory agents actively update external
memory through generated write, revise, and delete operations, but these updates may omit important information, corrupt existing
memory, or introduce unsupported hallucinated content. Once stored, such errors become persistent system-state failures that can affect future
reasoning and generation. In this paper, we propose TrustMem, a framework designed to improve the trustworthiness of
memory consolidation. TrustMem relies on a
Memory Transition Verifier to evaluate the transition process of
memory updates in terms of coverage, preservation, and faithfulness. It further constructs preference pairs among candidate updates under the same
memory state, enabling preference-guided reinforcement learning to directly optimize
memory updating behaviors. Extensive experiments demonstrate that TrustMem improves both
memory utility and reliability: it achieves state-of-the-art results across MemoryAgentBench, HaluMem, and the Mem-alpha validation set, improves HaluMem
memory extraction by 12.14 F1 points, and reduces transition-level omission, corruption, and
hallucination by 40.1%, 79.1%, and 50.0%, respectively, compared with the strongest baseline for each error type.
Read the full abstract ▾
All authors Tianyu Yang, Sudipta Paul, Vijay Srinivasan, Vivek Kulkarni, Srinivas Chappidi
Artificial Intelligence · Cross-listed from cs.LG
ATMA: Long-Context Language Modeling via Polar Attention and Gated-Delta Compression Memory
Habibullah Akbar
Length extrapolation in
language models involves competing objectives:
retrieval fidelity, long-document likelihood, short-context quality, and inference cost. We present ATMA, a 378M-parameter hybrid recipe that combines Polar
Attention with gated-delta recurrent
memory, and study these objectives as a Pareto problem rather than claiming general architectural dominance. Polar
Attention separates a normalized direction channel from a bounded participation-ratio magnitude channel. We select the recipe with a complete 120-
cell, 1B-
token factorial sweep, then train matched NoPE, RoPE, and Polar variants for 9.816B
tokens at length 2K and evaluate them through 256K. Across the factorial,
memory improves Polar's 64K
retrieval score in all 20 matched
cells (mean +47.8 points), whereas its effect on NoPE is small and inconsistent. At 256K, Polar retains 34.4% teacher-forced target-
token accuracy and 9.0% exact five-
token accuracy; exact
retrieval is 18.0% on synthetic contexts but 0.0% on FinePDFs contexts. Polar also limits mean fixed-target bits-per-byte degradation to 1.26 times, at a 1.9-point mean cost on eight short-context tasks. Raven baselines lead BABILong and have length-independent decode state, illustrating a different point on the frontier. Finally, a post-hoc checkpoint audit shows that nearly identical 2K validation curves can conceal a 6.70-nat difference at 256K. Because those runs were neither seed-paired nor randomized across devices, we interpret this as checkpoint variability associated with an infrastructure transition, not a causal hardware effect. Code: https://github.com/kreasof-ai/atma
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.CL
Hitting a Moving Target: Test-Time Adaptation for AI Text Detection under Continual Distribution Shift
Kevin Ren, Manish Raghavan and Nikhil Garg
Deployed AI text detection approaches are supervised, relying on training-time access to labeled
datasets of both human-written and AI-generated text. This approach is vulnerable to three types of distribution shifts that occur continually post-deployment, and for which labeled data is often unavailable:
adversarial humanization, new LLMs being released, and unidentifiable temporal drift in human writing after widespread LLM usage. Simultaneously, existing approaches do not leverage a key signal of LLM usage: inference-time homogeneity. We propose a test-time adaptation (TTA) approach that adapts to distribution shifts by leveraging homogeneity among unlabeled samples observed at inference time. Empirically, we find that state-of-the-art supervised detectors systematically fail when they encounter distribution shifts in AI-generated and human writing, both
adversarial and natural, while TTA with semi-supervised learning is largely robust; e.g., the commercial model Pangram detects just 27.0% of our
adversarial AI-generated text, compared to 93.3% for our test-time approach. We establish that TTA is a promising framework for AI text detection in the wild.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.LG
Silent Failures in Physics-Informed Neural Networks: Parameter Poisoning and the Limits of Loss-Based Validation
David McShannon and Nicholas Dietrich
Physics-informed neural networks (PINNs) embed governing equations in their loss function, enabling mesh-free solutions to partial differential equations. Low training loss is treated as evidence that the learned solution is physically correct. This paper shows that assumption breaks down when encoded physics are incorrect. By perturbing PDE parameters before training, a setting we describe as physics parameter poisoning or parameter misspecification, we produce models that train to low loss but give incorrect answers; we treat the perturbation schedule as sensitivity analysis rather than only as a security threat, and none of our claims requires an adversary. Achieving low residual loss does not discriminate accurate from inaccurate solutions: poisoned models reach losses at or below the clean baseline yet differ by large margins, so driving the residual down is not evidence of physical accuracy. Across three PDE systems (Burgers equation, Navier-Stokes cavity, and convection-
diffusion), poisoned models match or beat the clean-model training loss while their solutions differ by up to 71% in the fixed sweep and up to 128% under
adversarial search; at Cavity Re=400 the poisoned loss falls below the clean baseline. We define a detection difficulty ratio R (solution error divided by training loss) to summarize how invisible the corruption is, though cross-PDE comparison is complicated by differences in loss scale. We test six candidate defenses, none of which reliably detects corruption across all regimes. We propose a post-hoc defense: sweeping the PDE residual loss across parameter values without retraining. The loss minimum recovers the true training parameter without external data, and generalizes across all three PDE systems. The effect holds across five network architectures (8.7K to 133K parameters), is bidirectional, and is confirmed across multiple random seeds.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.HC
Proactive Systems in HCI and AI: Concepts, Challenges, and Opportunities
Nima Zargham, Sharon Ferguson, Jaisie Sin et al. (+2 more)
The last few years have seen a significant rise in interest in highly autonomous and proactive systems, fueled by advances in AI. Systems that anticipate user needs, take initiative, and act without explicit user input. Such systems span a wide range of applications, from smart lighting that adapts to user activity to assistive
robots that plan actions in advance to intelligent thermostats that learn routines and adjust environments proactively. Despite this breadth, the concept of proactivity remains loosely defined and inconsistently applied across research and practice. Current usage of the term often conflates fundamentally different system behaviors. For instance, simple reminders or recommendation systems are frequently labeled as proactive, even though underlying mechanisms and intentions differ significantly. This conceptual ambiguity limits our ability to systematically design, compare, and evaluate proactive systems. Moreover, existing methodologies for design and evaluation are largely rooted in reactive interaction paradigms, failing to address the unique challenges posed by proactive behavior, including timing, appropriateness, user control, transparency, and trust. This multidisciplinary workshop aims to establish a clearer and more rigorous foundation for understanding proactive systems. We bring together researchers and practitioners from Human-Computer Interaction, AI, and related fields to (1) develop a shared conceptualization of proactivity, (2) identify gaps and limitations in current design and evaluation approaches, and (3) co-create human-centered guidelines and research directions for future systems. Through interactive discussions and collaborative activities, the workshop seeks to map key challenges and opportunities, ultimately advancing robust and consistent frameworks for designing and evaluating proactive technologies.
Read the full abstract ▾
All authors Nima Zargham, Sharon Ferguson, Jaisie Sin, Cosmin Munteanu, Anastasia Kuzminykh
Artificial Intelligence · Cross-listed from cs.IR
TokenMinds: Pretrained User Tokens and Embeddings for User Understanding in Large Recommender Systems
Qingyun Liu, Bo Yan, Yang Liu et al. (+16 more)
User modeling in industrial recommender systems typically produces dense
embeddings, which suffer from representational constraints due to fixed-dimensional vectors. As an emerging alternative for discrete user representation, LLM-generated natural-
language user profiles often capture
language semantics rather than sequential patterns in user behaviors. Meanwhile, Semantic ID (SID) is effective for item representation, yet existing discrete user representations remain largely confined to quantizing already-pooled user vectors, clustering historical watches, or predicting coarse topic clusters solely for candidate
retrieval. We propose TokenMinds, an industrial system that extends the PLUM framework from SID-based item
retrieval to generative sequential user modeling. Built on an encoder-decoder architecture adapted from pretrained LLMs, TokenMinds introduces a dual-output design: the encoder outputs a dense user
embedding capturing a holistic user state, while the decoder generates discrete SID-based user
tokens representing multi-faceted future interests. Grounded in item semantics, these
tokens preserve distinct interests that a single pooled vector would average away. Additionally, a shared SID vocabulary across long- and short-form
videos unifies cross-scenario behaviors into a single model, cutting upstream training and serving compute by 50% and 31%. Furthermore, leveraging the text vocabulary inherited from the pretrained LLM, TokenMinds interleaves textual search queries directly with watch histories to capture explicit user intent. We validate TokenMinds through offline evaluations and live A/B experiments across YouTube's primary ranking surfaces, serving billions of users asynchronously: SID-based user
tokens are effective on their own, yield additive gains alongside dense
embeddings, and together outperform the previous-generation, non-LLM user model in production.
Read the full abstract ▾
All authors Qingyun Liu, Bo Yan, Yang Liu, Yuji Roh, Ekansh Sharma, Likang Yin, Emma Olowo, Yuxuan Li, Yikun Jiang, Diego Uribe, Saksham Aggarwal, Siqi Wu, Yuan Hao, Vikas Kedigehalli, Min-hsuan Tsai, Lukasz Heldt, Lichan Hong, Li Wei, Xinyang Yi
Artificial Intelligence · Cross-listed from eess.IV
Benchmarking the Alignment of Data-Quality Metrics, Human Judgment and Land-Cover Segmentation Performance for Earth Observation
Ümit Mert Çağlar and Alptekin Temizel
Volume and quality of
datasets are crucial for deep learning model training, yet they are often constrained by availability and data acquisition costs. Synthetic data augmentation can extend existing
datasets with realistic images, and the quality of these images is generally assessed through fidelity metrics such as FID, KID, IS, LPIPS and SSIM that measure structural or distributional similarity. However, such metrics, including the widely used FID, focus on visual fidelity without reflecting downstream utility, and can diverge from human perception under perturbations that are imperceptible to human observers. In this work, we systematically evaluate Earth observation
datasets alongside synthetic counterparts generated by deep generative models, comparing automatic metrics against human perception and downstream tasks. Our results reveal a stark misalignment: semantics-preserving perturbations such as rotation drastically alter metric scores while leaving human recognition unaffected, and synthetic samples that score poorly on automatic metrics achieve comparable or higher perceived realism, and can improve downstream performance when combined with real data. By benchmarking semantic
segmentation models trained on mixed real-synthetic
datasets, we demonstrate that quality metrics rooted in ImageNet-pretrained feature spaces are unreliable indicators for geospatial data. Our findings underscore that automatic quality evaluation of synthetic
datasets should be grounded in downstream task performance and human evaluation.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.LG
Reward-Conditioned Attention: How Reward Design Shapes What Autonomous Driving Agents See
Mohamed Benabdelouahad, Ahmed Djalal Hacini, Nadir Farhi et al. (+1 more)
We investigate how reward design shapes the internal
attention patterns of reinforcement learning
agents trained for autonomous driving. Using three Perceiver-based
agents that share identical architectures and training data but differ only in their reward configurations$\unicode{x2014}$ranging from basic violation penalties to continuous proximity penalties$\unicode{x2014}$we analyze cross-
attention allocation across 50 real-world scenarios from the Waymo Open Motion
Dataset. A central methodological finding is that naïve pooling of timesteps across episodes substantially underestimates the
attention$\unicode{x2013}$risk relationship; within-episode correlation with Fisher z-transform aggregation is the appropriate statistic and reveals a robustly positive link between collision risk and
agent-directed
attention. Building on this validated methodology, we demonstrate two reward-conditioned effects:
agents trained with
navigation rewards allocate up to $2.0\times$ more
attention to GPS-path
tokens than those trained with additional proximity penalties$\unicode{x2014}$and $4.7\times$ more than
agents with no
navigation incentive$\unicode{x2014}$revealing that reward content directly determines which scene elements the encoder prioritizes, and continuous time-to-collision penalties create a $\textit{learned vigilance prior}$$\unicode{x2014}$elevated resting
agent surveillance maintained throughout collision-free phases. In several scenarios, the complete-reward and minimal-reward models exhibit opposite
attention$\unicode{x2013}$risk correlation directions, demonstrating that reward design can qualitatively reverse attentional strategy rather than merely modulating its magnitude. These results suggest that
attention analysis is a practical diagnostic for verifying that a reward function produces the intended representational behaviour in safety-critical RL systems.
Read the full abstract ▾
All authors Mohamed Benabdelouahad, Ahmed Djalal Hacini, Nadir Farhi, Aissa Boulmerka
Artificial Intelligence · Cross-listed from cs.RO
AeroCast: Probabilistic 3D Trajectory Prediction for Non-Cooperative Aerial Obstacles via Transformer-MDN Architecture
Syed Izzat Ullah and Jose Baca
Autonomous aerial vehicles operating in shared airspace must predict the future positions of non-cooperative obstacles to plan evasive maneuvers before a collision becomes unavoidable. Unlike cooperative systems that share intent, non-cooperative obstacles such as birds, uncontrolled drones, or debris exhibit multi-modal motion that deterministic predictors cannot adequately represent. Existing methods either rely on recurrent encoders that propagate temporal information sequentially, limiting their ability to capture long-range kinematic precursors of maneuver initiation, or produce point forecasts that provide no distributional information to downstream planners. This paper presents AeroCast, a probabilistic trajectory prediction framework that combines a
Transformer encoder with a Mixture Density Network output head to predict per-timestep Gaussian mixture distributions over future three-dimensional displacements. A translation-invariant consecutive displacement encoding and a calibration-oriented training objective address the input design and mode-degeneracy challenges specific to mixture-based aerial trajectory prediction. On a hybrid real-and-synthetic quadrotor corpus spanning nine motion categories, AeroCast reduces Average Displacement Error and Final Displacement Error by approximately 50% relative to the baselines over a five-second horizon, and achieves the lowest negative log-likelihood and Continuous Ranked Probability Score among all compared methods. Ablation analysis identifies velocity input and model capacity as the primary contributors to prediction quality, and positional encoding as essential for long-horizon trajectory coherence. AeroCast inference completes in 0.1ms per sample, compatible with real-time onboard deployment at 100Hz.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from eess.AS
BCoughBench: Benchmarking Respiratory Acoustic Foundation Models Under Body-Coupled Wearable Sensor Conditions
Mayur Sanap, Prasanna Desikan and Edgar Lobaton
Respiratory acoustic foundation models (FMs) are benchmarked exclusively on smartphone recordings, yet clinical deployment increasingly targets body-coupled (BC) wearables whose sensors attenuate high-frequency content through tissue and bone, leaving FM reliability uncharacterised. We introduce BCoughBench, evaluating five FMs (OPERA-CT/CE/GT, HeAR, M2D+Resp) on nine classification tasks (AUROC, sensitivity at 95% specificity, Expected Calibration Error) and three age regression tasks (MAE vs. a mean-predictor baseline) across five EBEN-simulated BC sensor conditions on five labeled cough
datasets. Mean AUROC declines from 0.785 (smartphone) to 0.689-0.723, degrading most under temple vibration pickup ($\Delta$ = -0.096) and least under the soft in-ear ($\Delta$ = -0.062). No FM meets the clinical sensitivity threshold (Se@Sp95 $\geq$ 0.20) on most disease tasks under any BC sensor. Sex classification on the CIDRZ cohort collapses (AUROC 0.954 to 0.596-0.628, $\Delta$ = -0.341) while COVID detection is nearly unaffected ($\Delta$ = -0.004). Age regression is robust, improving under the forehead accelerometer on CoughVID (MAE 9.61 to 8.97 yr); HeAR leads on regression and demographic tasks, M2D+Resp on disease and characteristic tasks. BCoughBench provides a reproducible framework for FM evaluation under wearable conditions.
Read the full abstract ▾
Artificial Intelligence
The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
Eileanor LaRocco, Sarah Tan, Adarsh Subbaswamy et al. (+4 more)
Autonomous AI systems are transitioning from advisory roles to autonomous ones for medication prescriptions. Recent U.S. bill H.R. 238 and Utah's prescription-renewal pilot program both authorize AI to prescribe medications in an agentic capacity. While many regulatory guidelines suggest aggregate model performance metrics at the point of clearance, they do not require i) calibrated per-prediction confidence for action-gated thresholds, ii) differentiated communication between uncertainty arising from model ignorance (epistemic) from genuine clinical ambiguity (aleatoric), and iii) inferential transparency at the moment of decision enabling liability allocation. Here, we argue these three architectural features are minimum conditions for safe autonomous prescribing, and validate them with a survey of 136 U.S. prescribing clinicians. Our results suggest prescribing clinicians i) would not permit autonomous prescribing without a confidence-based escalation mechanism, ii) preferred a competing-options summary for aleatoric uncertainty but preferred abstention for epistemic uncertainty, and iii) were only willing to accept liability when inferential transparency enabled them to make a decision under acknowledged uncertainty. These findings indicate that our recommended architectural features would encourage higher rates of clinician adoption of autonomous AI prescribing, largely through collapsing much of what "autonomy" conventionally means.
Read the full abstract ▾
All authors Eileanor LaRocco, Sarah Tan, Adarsh Subbaswamy, Anne Andrews, Andrew Taylor, Cree Gaskin, Chirag Agarwal
Artificial Intelligence
Beyond Shapley: Efficient Computation of Asymmetric Shapley Values
Ezequiel Companeetz, Santiago Cifuentes and Sergio Abriola
We address the problem of explainability in machine learning models through feature attribution methods. In particular, we consider a variant of Shapley values known as Asymmetric Shapley Values (ASV), which enables the incorporation of causal knowledge into model-agnostic explanations through the use of a causal graph. We show that in certain contexts in which the computation of SHAP is $\#P$-hard, the exact computation of ASV can be done in
polynomial time. To extend this algorithmic result, we introduce a notion of equivalence classes over the topological orderings of the underlying causal graph, which is useful to reduce the time to compute ASV. In particular, we present a
polynomial-time
algorithm (in the number of equivalence classes) to compute it whenever the causal graph is a rooted directed tree. Finally, we develop an
algorithm for approximating ASV in arbitrary causal DAGs which relies on a procedure to sample topological orderings uniformly at random. To implement this sampling mechanism we leverage known
algorithms as well as simpler alternatives. Our experimental results demonstrate the practical viability of the proposed approach in realistic causal structures.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.DC
Power-Flexible AI Data Centers: A New Paradigm for Grid-Responsive Compute
Chris Williams, Philip Colangelo, Ayse Coskun et al. (+15 more)
The rapid expansion of artificial intelligence (AI) infrastructure is driving unprecedented growth in electricity demand from data centers. Traditional power-system planning treats large computing facilities as inflexible peak loads, leading to costly infrastructure upgrades and long delays in grid interconnection. Recent work has shown that AI clusters can reduce electricity consumption during peak demand through software-based workload orchestration. This article explores how modern GPU-based AI data centers can operate as grid-interactive assets that respond dynamically to power system conditions. We describe an architecture integrating grid signals, workload scheduling, and power telemetry for fine-grained cluster power control. Experimental results from a real-world deployment on a 130 kW GPU cluster demonstrate multiple forms of flexibility, including rapid load reduction, sustained curtailment, and carbon-aware operation while preserving service levels for priority jobs. We further demonstrate performance-aware load shifting across geographically distributed clusters, enabling workloads to migrate toward regions with lower grid stress. Together, these capabilities transform AI infrastructure from static electricity consumers into flexible resources that support grid reliability, accelerate interconnection, and improve computing sustainability.
Read the full abstract ▾
All authors Chris Williams, Philip Colangelo, Ayse Coskun, Ethan Levine, Andy Neale, Ciaran Roberts, Shayan Sengupta, Nikhil Shirolkar, Varun Sivaram, Sarah Soares, Ethan Tiao, Scott Underwood, Daniel Wilson, Frank Sharp, Luke Wainwright, Harry Petty, Scott Wallace, Brandon Records
Artificial Intelligence · Cross-listed from cs.LG
Training for the Model You Return: Improving Optimization for Iterate-Averaged Language Models
Kwok Chun Au and Adam Block
Many modern
Language Model (LM) pipelines return an averaged model, such as an exponential moving average of the training iterates, rather than the final iterate itself. This raises a fundamental question: given that we will return an iterate average, how should we change training to improve the performance of this average? We study this question by formulating optimizer design for the iterate-average
estimator as an optimal-control problem. In a continuous-time
stochastic quadratic model, we solve for the control strategy that minimizes the error of the returned average subject to a penalty on the size of the intervention. A practical approximation to this controller yields PACE, a lightweight wrapper around AdamW that pulls the live weights toward their exponential moving average with a clipped, per-coordinate control strength. We prove that a stylized version of PACE converges at the standard
stochastic convex optimization rate, up to a factor depending on the averaging rule, while in the quadratic setting it can strictly improve the limiting squared error of the iterate-average
estimator and can do so by an arbitrarily large factor on some instances. Empirically, our results suggest that PACE improves over AdamW and EMA-evaluated AdamW in supervised fine-tuning of 1-2B parameter LMs and in GPT-2 pretraining on FineWeb for a wide range of learning rates, decay schedules, and other hyperparameters.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.LG
GCT-MARL: Graph-Based Contrastive Transfer for Sample-Efficient Cooperative Multi-Agent Reinforcement Learning
Animesh Animesh, Satheesh K Perepu and Kaushik Dey
In cooperative multi-
agent reinforcement learning (MARL), from a deployment perspective, it is challenging and expensive to train
agents from scratch for each new environment or task. In this work, we propose GCT-MARL, a transfer learning framework that builds on the multi-view graph contrastive backbone of MAIL and augments it with a per-view, adaptively weighted
alignment loss and a two-phase training protocol specifically designed for transfer across populations of varying sizes and compositions. We empirically demonstrate that the proposed framework markedly accelerates convergence on the target task relative to from-scratch training, in both homogeneous (within-faction, varying N) and heterogeneous (cross-faction and mixed unit-type) transfer scenarios. Furthermore, we show that the framework naturally supports continual learning by sequentially chaining the two-phase transfer protocol across a series of related tasks. Overall, this work provides a unified approach to mitigating key limitations in current MARL transfer methods with new insights at both methodological and empirical levels.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.LG
Adapt Only When It Pays: Budgeted Decision-Loss Priority for Delayed Online Time-Series Adaptation
Xibai Wang
Online time-series forecasters receive labels only after horizon-dependent delays, while every adaptation step spends limited compute. We study when an online learner should update, not how to adapt at every opportunity, and introduce ADOWIP: a residual-adapter framework with sealed delay queues, exact budget accounting, and auditable update telemetry. Its main scheduler is an observed decision-loss priority gate that updates only after feedback is revealed, when downstream loss, optionally penalized by prediction MSE, exceeds a calibrated empirical
quantile and budget remains. We prove hard-budget feasibility, projected-OGD regret for a convex linear accepted-update subproblem, and stability plus conditional finite-sample gate-selection statements. On public ETT capacity-planning tasks, a frozen calibration/evaluation split selects a gate that lowers held-out decision loss against always, fixed-period, and drift-triggered exact-update baselines under matched compute. Secondary threshold/load-index ETT suites are mixed: 33 of 41 selected contrasts clear the stricter cross-artifact Holm family, and the 8 nonpassing rows are explicitly excluded from primary claims. The same protocol improves an external UCI Bike capacity proxy with 20/0 held-out wins, and a fixed gate passes three full-year Capital Bikeshare station-rebalancing contrasts. Probe-based and finance experiments remain negative, delimiting the current scope of decision-prioritized adaptation.
Read the full abstract ▾
Artificial Intelligence
Do vision-language models search like humans? Reasoning tokens as a reaction-time analog in classic visual-search paradigms
Farahnaz Wick
Visual search has been one of the most productive paradigms in the study of visual
attention: the way reaction time scales with the number of items distinguishes parallel, "pop-out" search from serial,
attention-demanding search. I ask whether
vision-
language models (VLMs) exhibit the same behavioral signatures. I adapt four classic paradigms: feature versus conjunction search, spatial-configuration (T-vs-L) search, enumeration, and the tilted/vertical search asymmetry; and present them to current frontier and mid-tier models. Because a single model call has no reaction time, I use the number of
reasoning ("thinking")
tokens a model spends per trial as a within-model analog of search effort, and I compare against a large public human
benchmark (Wolfe et al., 2010). The models reproduce several human signatures: feature search costs flat effort while conjunction effort climbs with set size; frontier models hold accuracy where mid-tier models collapse to chance; and a resolution control shows the conjunction cost is genuine search rather than difficulty resolving small shapes. They also diverge from humans in informative ways. The target-present effort slope exceeds the target-absent slope, reversing the human ordering; enumeration remains accurate where humans would lose count; and a
reasoning model with adaptive deliberation declines to deliberate on detection tasks altogether, so that a single search expresses itself as an effort gradient in one model and as an accuracy cliff in another. I argue that psychophysical paradigms, applied behaviorally, are a sharp and inexpensive probe of machine visual cognition, and that the points of divergence are as informative as the points of agreement.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.CR
What Does It Mean to Break a Distillation Defense?
Lena Libon, Pura Peetathawatchai, Michael Aerni et al. (+2 more)
Black-box LLMs (accessible only via API) are vulnerable to
distillation attacks, in which an attacker queries the model and trains a student on its outputs. A recent line of work proposes output perturbation defenses that modify the teacher's output to reduce student performance while preserving utility for legitimate users. As a relatively new family of approaches, output perturbation defenses lack a shared threat model, making it difficult to compare them, reason about composing them with other attacks, or evaluate their robustness against realistic adversaries. This underspecification matters beyond technical evaluation: when defenses are deployed to protect intellectual property or justify regulatory compliance, an imprecise threat model can create a false sense of security. We propose a threat model framework that describes attackers along three dimensions: a query budget, a data budget, and an interface profile that captures how attackers interact with the API. Using antidistillation sampling as a case study, we show that whether the defense is considered effective depends on the assumed threat model. We argue that future work on
distillation defenses, along with any governance or policy frameworks built around them, should explicitly specify and stress-test attacker capabilities along our three dimensions.
Read the full abstract ▾
All authors Lena Libon, Pura Peetathawatchai, Michael Aerni, Daniel Paleka, Florian Tramèr
Artificial Intelligence · Cross-listed from cs.CV
Wan-Streamer v0.1: End-to-end Real-time Interactive Foundation Models
Lianghua Huang, Zhi-Fan Wu, Wei Wang et al. (+22 more)
We present Wan-Streamer, a native-streaming, end-to-end interactive foundation model designed from the ground up for real-time, low-latency, full-duplex audio-visual interaction. Wan-Streamer seamlessly models
language, audio, and
video as both input and output within a single
Transformer, where the sequence is represented as interleaved visual, audio, and text input
tokens together with visual, audio, and text output
tokens, coordinated by block-causal
attention for incremental streaming. Unlike cascaded interactive systems that rely on separate VAD, ASR,
language, TTS, audio-driven animation, or
video-generation modules, Wan-Streamer does not rely on external
language,
speech, avatar, or
video-generation modules: perception,
reasoning, generation, response timing, turn management, and cross-modal synchronization are learned jointly within one unified model, reducing pipeline latency and error accumulation. To support natural audio-visual responsiveness, we redesign the entire stack around streamability, including causal encoders, causal decoders, block-causal
attention, and low-latency
multimodal token scheduling, enabling streaming units as short as 160 ms at 25 fps. Wan-Streamer achieves approximately 200 ms model-side response latency and approximately 550 ms total interaction latency when combined with 350 ms bidirectional network latency, supporting sub-second duplex audio-visual communication. These results position Wan-Streamer as a unified, end-to-end,
multimodal interactive foundation model for low-latency streaming interaction.
Read the full abstract ▾
All authors Lianghua Huang, Zhi-Fan Wu, Wei Wang, Yupeng Shi, Mengyang Feng, Junjie He, Chen-Wei Xie, Yu Liu, Jingren Zhou, Ang Wang, Bang Zhang, Baole Ai, Chen Liang, Cheng Yu, Chongyang Zhong, Jinwei Qi, Kai Zhu, Pandeng Li, Peng Zhang, Wenyuan Zhang, Xinhua Cheng, Yitong Huang, Yun Zheng, Yuzheng Wang, Zoubin Bi
Artificial Intelligence · Cross-listed from cs.LG
LLM-ACES: Closed-Loop Discovery of Dynamical Systems with LLM-Guided Adaptive Search
Nikhil Abhyankar, Sha Li, Sanchit Kabra et al. (+3 more)
Recovering governing Ordinary Differential Equations (ODEs) from data is a central challenge in modeling dynamical systems across scientific domains. Existing approaches cast discovery as a static inference problem over fixed
datasets, assuming that the observed trajectories are sufficiently informative. However, dynamical systems evolve over large state spaces, and limited data can make multiple equations observationally indistinguishable, leading to identifiability gaps and the recovery of incorrect governing equations. To address this, we introduce LLM-ACES, or LLM-guided Active Closed-loop Equation Search, a closed-loop framework that jointly optimizes symbolic hypothesis construction and adaptive data acquisition. In LLM-ACES, a large
language model (LLM) proposes operator priors that partition the large search space into distinct regions, within which candidate equations are fit to the observed data. The disagreement among these candidates guides the acquisition of informative trajectories, creating a feedback loop that iteratively refines both the hypothesis space and the discovered dynamics. On 122 ODE systems spanning ODEBench and ODEBase, LLM-ACES achieves the lowest median NMSE, outperforming state-of-the-art baselines by several orders of magnitude while achieving a high symbolic accuracy of 46.2% and 52.4%, respectively. Our analysis further shows that LLM-ACES is sample-efficient, achieving better performance with one-tenth the data. Furthermore, LLM-ACES's feedback-driven data acquisition makes it robust to noise and recovers the correct symbolic structure, while baselines introduce spurious terms that fit the data locally but obscure the true governing relationships.
Read the full abstract ▾
All authors Nikhil Abhyankar, Sha Li, Sanchit Kabra, Naren Ramakrishnan, Yulia Gel, Chandan K. Reddy
Artificial Intelligence · Cross-listed from cs.CV
Yuvion VL: A Multimodal Foundation Model for Adversarial Content and AI Safety
Shikai Qiu, Xiaowen Xu, Benlei Cui et al. (+55 more)
General-purpose models often struggle to reliably identify and understand real-world
multimodal risks, largely due to the inherent
multimodal adversarial nature of content and AI safety. We present Yuvion VL, a family of
multimodal large
language models purpose-built for content and AI safety, with both instruction-tuned and
reasoning-oriented variants. Yuvion VL addresses this gap by treating safety as an inherently
adversarial and
multimodal problem and designing the entire pipeline around
adversarial robustness. For data construction, we develop an automated pipeline integrating
adversarial-aware data synthesis with multi-stage quality control, producing large-scale, high-quality
multimodal samples augmented with domain knowledge and
reasoning annotations. For training, we adopt a three-stage pipeline that includes continued pretraining for risk-concept cross-modal
alignment, instruct post-training for production-grade safety tasks, and
reasoning post-training for enhanced interpretability and performance in complex tasks. We further introduce Confuse-then-Contrast Fine-Tuning, a contrastive framework that mines model-specific confusions and constructs multi-image contrastive groups to enforce explicit discrimination of fine-grained visual-semantic elements, enabling the model to distinguish between visually similar cases with different safety implications in
adversarial safety tasks. To support rigorous evaluation, we further introduce Yuvion VL RiskEval (YVRE), a collection of
benchmarks covering diverse open and internal evaluations, with a focus on content and AI safety,
adversarial robustness, and real-world capability requirements. Experiments show that Yuvion VL-32B achieves industry-leading safety performance, surpassing comparably sized open-source models and best closed-source commercial models, while maintaining comparable general capabilities.
Read the full abstract ▾
All authors Shikai Qiu, Xiaowen Xu, Benlei Cui, Ting Ma, Xiufeng Huang, Wenjing Jiang, Shaoxuan He, Haolei Xu, Chunyang Chai, Yujian Li, Yiliang Zhang, Guanghui Wang, Ziheng Wang, Ziwen Xu, Zhaoyu Fan, Jinhao Chen, Ruijie Jian, Hongxing Li, Chuxi Xiao, Xinyue Chen, Wenxuan Liu, Libin Dong, Yupeng Cao, Xiaoqian Xia, Jing Wang, Zhe Jiang, Zhenan Ye, Guang Yang, Bin Liu, Wei Peng, Ziqiang Zhu, Meihui Lian, Kaiwen Lv Kacuila, Haidong Ding, Dongjie Zhang, Yangfan Zhou, Bingyu Zhu, Yan Wang, Hai Zhao, Xuan Jin, Wei Zhao, Pengfei Sun, Huiming Zhang, Wei Wang, Xipeng Cao, Jialun Chen, Xiao Chen, Shaola Ren, Yunqing Hu, Bin Li, Chengwen Yao, Meng Huang, Xianfeng Li, Bin Tang, Chao Liu, Hui Xue, Longtao Huang, Haiwen Hong
Artificial Intelligence · Cross-listed from cs.LG
Do Thinking Tokens Help with Safety?
Narutatsu Ri, Abhishek Panigrahi and Sanjeev Arora
Today's
reasoning models use thinking
tokens to attain stronger performance on
benchmarks than their instruction-tuned counterparts. It is also generally believed that this more "deliberative" mode should improve
alignment and safety, by providing the model a safe space to consider whether its planned answer to a request violates its safety principles. We present evidence that this intuition is not always correct. Across frontier open-weight
reasoning models spanning GPT-OSS, Qwen, Olmo, and Phi families, we find that the eventual refusal/compliance outcome is already strongly predictable via a trained head on the first
token's hidden representation ($0.84$-$0.95$ AUROC and $\sim88\%$ balanced accuracy for predicting refusal/compliance) before any visible thinking. The thinking process turns out to be more akin to prefix completion than to deliberative revision, with the final outcome rarely changing after the first $\sim20\%$ of thinking, despite giving the appearance of deliberation at the text level ($\sim74\%$ of text-level deliberations occur when the response distribution is already locked to one refusal/compliance side). We also find that existing inference-time and training-based safety interventions, despite being motivated by the goal of inducing deliberation, largely shift model behavior toward over-refusal while suppressing already-scarce deliberation signals. Our results suggest that safety behavior in current
reasoning models is much less deliberative than commonly assumed, and highlight the need for methods that induce real safety deliberation.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.CV
Noise-Aware Boundary-Enhanced Generative Learning for Ultrasound Speckle Reduction
Yuexi Gu, Mengqi Wu, Yongheng Sun et al. (+3 more)
Ultrasound is a non-invasive, real-time, and cost-effective imaging technique widely used in clinical diagnosis. However, its diagnostic efficacy is often compromised by inherent speckle noise that degrades image quality and obscures underlying anatomical structures. Existing speckle reduction methods tend to over-smooth tissue boundaries and generalize poorly to heterogeneous noise levels. To address these limitations, we propose a Noise-Aware Boundary-Enhanced Generative Learning (NBGL) framework for ultrasound speckle reduction, which simultaneously preserves annotated anatomical boundaries and adapts to varying noise levels. The NBGL framework consists of a speckle reduction branch and a boundary enhancement branch. The former leverages generative learning to suppress speckle noise, while the latter learns boundary-sensitive representations to preserve target anatomical structures. Furthermore, a noise-aware interaction weight generation (NIWG) module estimates the speckle noise level via 3D
Laplacian filtering and a median absolute deviation
estimator, and translates it into an adaptive interaction weight. This weight is incorporated into a weighted feature-wise linear modulation (wFiLM) module to adaptively modulate cross-branch feature coupling, thereby improving robustness to varying noise levels. Extensive evaluations on 141 3D transvaginal ultrasound volumes demonstrate that NBGL consistently outperforms state-of-the-art methods in speckle reduction and structural preservation across six noise levels, while maintaining consistency with annotated anatomical boundaries.
Read the full abstract ▾
All authors Yuexi Gu, Mengqi Wu, Yongheng Sun, Virginie Papadopoulou, Mingxia Liu, Maureen Kohi
Artificial Intelligence · Cross-listed from cs.LG
Erased, but Not Gone: Output Forgetting Is Not True Forgetting
Teresa Pui Yee Yong, Win Kent Ong and Chee Seng Chan
Machine unlearning (MU) is commonly judged by output forgetting, such as low forget-set accuracy or reduced membership inference. But if output-level success can coexist with retraining-inconsistent residuals in representation space, what kind of forgetting are current evaluations actually certifying? We study this question through retraining-consistent representation forgetting, using a population of retrained models (i.e., trained from scratch without the forget data) as an operational reference distribution for ideal forgetting. Across multiple unlearning methods,
datasets, and models, our theoretical analysis and empirical results show that standard output-level evaluation can systematically overestimate the success of unlearning. Under this stronger lens, current methods often appear forgotten at the output layer while exhibiting a structured mismatch relative to retraining. They partially align with retraining on forget samples, remain more inconsistent on retain samples, and tend to leave residual discrepancy localized along retraining-related directions rather than diffuse in representation space. This structured mismatch is characterized by forget/retain asymmetry, directional mismatch, and concentrated residuals along retraining-related directions. In many cases, the discrepancies exceed the variability of independent retraining and lead to a risk of information recovery during relearning attacks. Overall, the results suggest that current MU is often evaluated for apparent forgetting rather than retraining-consistent forgetting. More broadly, retraining reveals what output forgetting hides.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.LG
Geo-Strat-RL: Learning Geological Event Reasoning from Verifiable Tasks
Lukas Mosser
To evaluate whether
vision-
language models can reason about geological histories, it is necessary to construct observations for which the underlying process history is known. Furthermore,
reasoning over geological histories is not just a question of recognizing visual patterns, but also of understanding temporal and structural relationships that may be only indirectly visible or highly ambiguous. When ground-truth event histories are not uniquely identifiable or are unavailable, it remains an open challenge to teach models capable of visual
reasoning to produce valid geological reconstructions that are consistent with both observed evidence and geological principles. We therefore investigate whether defining a verifiable geological
reasoning task can improve geological event reconstruction across observation domains through reinforcement learning with verifiable rewards (RLVR). To this end, we present Geo-Strat-RL, a synthetic environment that generates stratigraphic observations and compact visible-evidence event histories. The environment combines a geological generator with an executable verifier that scores chronology, event identity, deposition, and structural relationships. We show that RLVR improves geological reconstruction in
vision-
language models (VLMs), increasing geological content scores on held out stratigraphic diagrams. We further evaluate the same held-out geological histories in a synthetic seismic observation domain by converting the generated scenes into acoustic-impedance-derived amplitude sections. In this controlled paired-renderer setting, we present evidence that geological
reasoning learned from stratigraphic diagram-domain RLVR training transfers to synthetic seismic representations without seismic-specific training examples, supporting the hypothesis that RLVR can teach reusable geological
reasoning concepts across related observation formats.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.LG
Internal Data Repetition Destroys Language Models
Jessica Chudnovsky, Joshua Kazdan, Noam Levi et al. (+6 more)
Language models are running out of high-quality training data, and even aggressively deduplicated corpora retain some amount of repetition. Earlier controlled studies predated Chinchilla-style scaling laws and could only measure the cost of repetition indirectly. We revisit repetition in the Chinchilla era, using a fitted no-repetition scaling law to report Compute-Equivalent Gain and Compute-Equivalent Loss. We show that under this modernized paradigm, repetition damage is systematic in three ways. First, holding compute allocated to repeated data constant, eval loss peaks at an intermediate repeat count $\Rep$; repeating a moderately sized subset a moderate number of times damages performance more than repeating a large subset a few times or a small subset many times. Second, the location of this peak is well-fit by a power law in model size; this scaling law reveals that the most damaging number of repeated data grows more quickly than compute. Finally, when repeated documents consume 10% of the FLOPs budget in a controlled exact-document repetition setting, the compute-equivalent loss can be large: on FineWeb-Edu-Dedup, the most damaging repeat count for a Qwen3-style 344M-parameter model at $\OT=1$ matches the loss of a no-repetition run using 67% of the FLOPs. We demonstrate that these phenomena are not
language-model-specific, and can be analytically understood in a simple statistical model: a misspecified linear regression with verbatim duplicates reproduces the same qualitative loss peak, quantifying how such peaks can arise from a statistical tradeoff between memorization and generalization. Our findings add precision to the study of duplication in
language models, allowing practitioners to quantify the wasted compute incurred by the presence and repeat structure of duplicates in pretraining corpora.
Read the full abstract ▾
All authors Jessica Chudnovsky, Joshua Kazdan, Noam Levi, Rylan Schaeffer, Yegor Denisov-Blanch, Bo He, Mehmet Donmez, Sanmi Koyejo, David Donoho
Artificial Intelligence · Cross-listed from cs.LG
Are Tabular Foundation Models Robust to Realistic Query Distribution Shifts in Microbiome Data?
Giulia Perciballi, Ahmad Fall, Federica Granese et al. (+2 more)
Tabular foundation models (TFMs) achieve strong performance on microbiome abundance data, yet their robustness under realistic distribution shift remains poorly characterized. We introduce a
benchmark that evaluates the robustness of TFMs to biologically inspired perturbations across six gut microbiome
datasets spanning four disease contexts. In this in-context learning setting, models receive unperturbed support sets as context and are evaluated on perturbed query samples. To isolate robustness beyond "shortcut" features, we preserve the most discriminative taxa and apply three controlled perturbation strategies: (i) removal of high-abundance (uninformative) taxa, (ii) sparsification via increased zero-inflation, and (iii) zero-imputation via spurious non-zero injections. Our results show that protecting discriminative features is insufficient to guarantee stability under support-query shift: across
datasets, all perturbations degrade model performance, with zero-imputation consistently the most harmful, indicating that corrupting global feature structure can break generalization even when key taxa are retained. Sparsification disproportionately affects TFMs relative to a classical random forest baseline, suggesting greater sensitivity to zero-inflation-type shifts. The code is publicly available at: https://github.com/UMMISCO/metagenomics-fm/.
Read the full abstract ▾
All authors Giulia Perciballi, Ahmad Fall, Federica Granese, Edi Prifti, Jean-Daniel Zucker
Artificial Intelligence · Cross-listed from cs.LG
ExTra: Exploratory Trajectory Optimization for Language Model Reinforcement Learning
Wenyang Hu, Junxiang Jia, Zhen Shu et al. (+3 more)
Reinforcement Learning with Verifiable Rewards (RLVR) for
language-model
reasoning can fail at both extremes of task difficulty: easy
prompts often produce all-correct, low-diversity rollout groups with little gradient signal, while hard
prompts can produce all-incorrect groups with no positive reward. We introduce ExTra (Exploratory Trajectory Optimization), a GRPO-compatible framework that extracts exploration signals from the model's own rollouts. ExTra combines two mechanisms: (i) a novelty reward that adds
embedding-based diversity bonuses after GRPO normalization, rewarding diverse correct solutions; and (ii)
entropy-guided prefix regeneration, which scores partial trajectories using
entropy signals and continues exploration from promising intermediate steps. Across six mathematical
reasoning benchmarks, ExTra improves Qwen3-1.7B over GRPO by about +5 points on pass@1 and +7 points on pass@16, showing that trajectory-level exploration signals can improve both single-sample accuracy and inference-time coverage.
Read the full abstract ▾
All authors Wenyang Hu, Junxiang Jia, Zhen Shu, Daniel Dahlmeier, See-Kiong Ng, Bryan Kian Hsiang Low