Peer Revenue Fully funded

Papers

Fresh from arXiv, typeset properly. Highlighted words trade on the market.

Computer Science Artificial Intelligence

cs.AI Watch Artificial Intelligence

Monday 5 October

Artificial Intelligence

AMBER: Training Long-Horizon Web Agents through Append-Only Memory

Chinmay Savadikar, Zhaoyu Zhang, Mingyu Zhao et al. (+4 more)

Modern language-model agents increasingly interact with external environments over long-horizon, multi-step trajectories, where the accumulated interaction history can quickly exceed practical context budgets. To ensure reliability, agents must maintain factual information over long horizons, remember execution errors and corrective feedback, and track progress across actions. Several approaches have been proposed to achieve this without the need for maintaining the entire execution history in context, such as using the reasoning and action history, learning to maintain a fixed-size memory through an overwrite mechanism, and periodic summarization. Although overwrite memory can in principle retain anything an append-only memory can, it must learn to carry each fact through every subsequent rewrite, which is difficult to learn from sparse outcome rewards; for interactive applications like web agents, we find that trained overwrite memories delete key information required by the trajectory, as well as corrective feedback received from the environment. We introduce AMBER (Append-only Memory Bank for Evidence Retention) - a simple and scalable framework where an agent jointly learns to reason, act, and write free-form memory, while an append-only rule guarantees retention by construction. This allows AMBER to be trained end-to-end with reinforcement learning from outcome rewards without the need for extensive curated SFT data. On WebArena Lite, AMBER improves average success over overwrite-based memory by 4.09 percentage points, increases the fraction of tasks solved in five repeated runs by 4.8 percentage points, and matches an overwrite baseline trained on substantially more expensive curated supervision. AMBER achieves these improvements while maintaining a practical token budget, providing a strong balance between context efficiency, task performance, and reliable long-horizon execution.
Read the full abstract

All authors Chinmay Savadikar, Zhaoyu Zhang, Mingyu Zhao, Shuang Xie, Han Li, Tianfu Wu, Lingyun Wang

Artificial Intelligence · Cross-listed from cs.LG

Will the Judge Flip? Predicting Position-Sensitive LLM Judgments from Residual Stream Activations

Hashmath Shaik, Gnaneswar Villuri and Alex Doboli

The order in which candidate responses are presented can change an LLM judge's verdict. Detecting such a position flip ordinarily requires judging each pair in both orders, which doubles the number of judgments. We investigate whether residual stream activations recorded immediately before the initial verdict can predict a flip. We use nested grouped cross-validation to evaluate regularized linear probes on 534 JudgeBench pairs for three Qwen3 judges and Llama-3.1-8B. The linear probes achieve AUROCs of .621-.850 and outperform a combined baseline that uses verbalized confidence, verdict-label logits, response lengths, and the judge's initial choice by .062-.113 AUROC. Linear probes trained on JudgeBench and then frozen achieve AUROCs of .685-.853 on 1,802 MT-Bench comparisons without MT-Bench fitting or recalibration. These results show that pre-verdict activations support prediction of susceptibility to candidate order and outperform the non-activation predictors evaluated here.
Read the full abstract

Artificial Intelligence · Cross-listed from cs.LG

LiLib: Lifelong Air-to-Ground Path-Loss Prediction on UAVs via a Drift-Triggered Model Library

Minh Tran

UAVs that act as relays or base stations need accurate air-to-ground path-loss predictions for rate adaptation and placement, but propagation conditions change as a UAV moves between suburban, urban and high-rise areas, and the same areas are often revisited. Online regressors that adapt by forgetting must relearn each environment from scratch, whereas a single model trained on all data averages incompatible regimes. We propose LiLib, a lightweight continual-learning scheme in which a UAV maintains a small library of recursive-least-squares experts. A windowed residual test detects drift; a short probe phase then either reuses the best stored expert or creates a new one. In simulations based on four standard urbanization profiles, LiLib reduces prediction RMSE from 5.89 dB (best sliding-window baseline) to 4.03 dB (p < 0.001), lowers the error shortly after a return to a known environment from 12.3 dB to 5.7 dB, and recovers 99% of the throughput of a regime-aware oracle in rate adaptation. The library stores four experts in under 0.5 KB, and identifies regimes with 92% purity without labels. When a second UAV is initialized with the library of a peer, its error after environment changes halves. LiLib does not reach the oracle, and similar regimes may be merged when shadowing is strong. The results indicate that, for recurring drift, remembering is more effective than re-adapting.
Read the full abstract

Artificial Intelligence · Cross-listed from cs.IR

Beyond Successor Accuracy: State Retention for Recursive Self-Improvement in Recommendation

Jinfeng Xu, Zheyu Chen, Ziyue Peng et al. (+5 more)

Recommendation recursive self-improvement (Rec-RSI) feeds recommender outputs into subsequent training. Evaluating each round solely through its latest model assumes that the successor consolidates the update, although pre- and post-update models may retain complementary ranking decisions. We term this distributed progress and quantify it using cross-generation advantage (CGA), a marginally matched contrast between cross- and within-generation model pairs. A rank-separation statistic, label-free at selection time, predicts which family to retain. Across four datasets and three sequential recommendation encoders, the preferred retention regime varies by architecture: cross-generation pairing benefits GRU4Rec and SASRec, whereas FMLP initially favors within-generation pairing and shifts toward cross-generation pairing after a second update. Rank separation selects the stronger family in 12/12 first-update and 5/6 second-update dataset-encoder settings; on held-out tests, the selected family outperforms the direct successor in 34/36 trajectories. Five transfer mechanisms do not consistently reproduce these gains in one model. These findings establish state retention as a distinct Rec-RSI problem: progress may reside in relations between generations as well as in the latest model. Code is available at \href{https://github.com/Jinfeng-Xu/RecRSI}{https://github.com/Jinfeng-Xu/RecRSI}.
Read the full abstract

All authors Jinfeng Xu, Zheyu Chen, Ziyue Peng, Zheng Lin, Wenhao Yuan, Jian Chen, Shujie Li, Edith Ngai

Artificial Intelligence · Cross-listed from cs.LG

Muon Is Theoretically Wrong For Convolutions, But Empirically Effective

Thibaut Boissin, Thomas Massena, Mathieu Serrurier et al. (+1 more)

Muon, an optimizer known for its efficiency, has a clear interpretation for matrix-valued updates, but convolutional kernels are stored as four-dimensional tensors. Standard implementations reshape these tensors into matrices, a shortcut which breaks the theoretical understanding behind Muon. To investigate this, we formalize the corresponding optimization objective directly in convolutional operator geometry and introduce Convolutional Newton-Schulz (Conv-NS), which approximates the polar factor in this geometry while preserving kernel support. When applied in fast training experiments, Conv-NS and reshape-based Muon are both computationally efficient and achieve comparable accuracy on CIFAR-10 and ImageNet classification tasks. However, as one could expect a theoretically aligned Conv-NS to outperform reshape-based Muon, we investigate this mismatch between practice and theoretical understanding, with the hypothesis that exact convolutional orthogonalization may overconstrain updates. These findings highlight Muon's strong practical performance while opening directions for its further development on convolutions. Our code is publicly available at \href{https://github.com/thib-s/muonconv-cifar10-airbench}{github conv-muon}.
Read the full abstract

All authors Thibaut Boissin, Thomas Massena, Mathieu Serrurier, Franck Mamalet

Artificial Intelligence

When to Remember, When to Abstain: Category-Conditioned Retention for Reliable Agent Memory

Olukunle Owolabi, Pulkit Gupta and Fei Wang

Persistent agent memory is only as reliable as its retention decision: an assertion weakly supported by its source can be stored and later reused as established fact. We study whether the retention decision should be governed by a confidence bar conditioned on the semantic category of the assertion rather than by a single global threshold, retaining well-evidenced categories liberally while abstaining more aggressively where inference is unreliable. We evaluate this in a deployed cold-start memory pipeline on 100 synthetic personas. The empirical evaluation is motivated by a sharp reliability asymmetry: across 4{,}715 candidate assertions, only 77.9% of value and belief assertions are supported by their source, versus 96.2% for all other categories. A global confidence threshold cannot separate these: it either admits unsupported value claims or discards well-evidenced ones. Conditioning the threshold on category resolves the tradeoff. In repeated held-out evaluation, a stricter bar on values alone reduces unsupported retentions from 6.2% to 4.0% (an ${\approx}36\%$ relative reduction, modest but consistent across folds) and, as corroborating evidence, preserves an estimated 13 percentage points more coverage (95% CI 9.8–16.0) than a global threshold at comparable retention. Our results suggest that reliable retention depends on the type of assertion, not on confidence alone, and that a category-conditioned threshold can act as a simple, effective form of selective prediction at the write boundary.
Read the full abstract

Artificial Intelligence · Cross-listed from cs.DC

T-CCL: Resource Efficient and Performant Collective Communication using Tensor Memory Accelerator

Keyvan Dadashzadeh, Yuehong Zhou, Minyu Cui et al. (+1 more)

Large transformer-based models increasingly depend on multi-GPU execution, which requires frequent collective communication among GPUs. Existing communication libraries often rely on many GPU threads to achieve high bandwidth or low latency, resulting in a large streaming multiprocessor (SM)-side resource footprint. This footprint can limit the resources available to other GPU work, particularly when communication and computation execute concurrently. Thus, efficient collective communication should not only achieve high collective performance but also reduce its SM-side resource usage. This paper presents T-CCL, a resource-efficient collective communication library based on the Tensor Memory Accelerator (TMA) for intra-node communication. T-CCL offloads both data movement and reduction operations to TMA and executes each collective as a pipelined series of asynchronous TMA operations, reducing the SM resources required for collective communication while maintaining high bandwidth. Evaluated across AllReduce, AllGather, and ReduceScatter collectives, T-CCL outperforms NCCL by up to 2.4x with unrestricted communication resources and up to 3.42x under restricted resource budgets, remains competitive with NCCL's recent symmetric-memory kernels, and occupies the same or fewer SMs in profiled cases. In a GEMM-collective overlap case study, switching the communication backend from NCCL to T-CCL raises the average operator-level speedup over a sequential baseline from 1.12x to 1.25x on two GPUs and from 1.04x to 1.14x on four GPUs, as T-CCL uses fewer SMs for communication, leaving more SMs available to the overlapped GEMM. Integrated into vLLM as a communication backend, T-CCL improves end-to-end inference throughput over vLLM's automatic backend dispatch by up to 1.31x, outperforming it at every evaluated batch size on both the conversation and decode-heavy workloads.
Read the full abstract

All authors Keyvan Dadashzadeh, Yuehong Zhou, Minyu Cui, Miquel Pericas

Artificial Intelligence

Verified, not generated: expert-verified AI study materials and the distribution of learning gains in a university course

Canh Thien Dang and An Nguyen

Experimental studies of generative AI in education mostly report average effects, yet field evidence shows that AI can narrow attainment gaps or widen them. We argue that the direction depends on the judgement burden, the expertise a learner must supply to screen AI output before learning from it, and that expert verification before release moves this burden from students to an accountable tutor. We test the argument in a two-cohort difference-in-differences design in which one half of a compulsory firstyear university economics course received AI-generated podcasts, FAQs and quiz-based study guides, produced with a source-grounded model and checked by a named graduate teaching assistant (170 students; 340 examination marks). Access was associated with a 2.34-mark advantage on a 50-mark component. The share of marks below the upper-second classification boundary fell by 24.7 percentage points relative to the counterfactual, effects were significant at every threshold from 23 to 31 marks and at none above, and roughly three-quarters of the average originated in the bottom quintile. The threshold estimate is robust to removing the lowest-scoring students from the pre-intervention cohort; the average effect is not. Interviews and feedback from 36 students indicate that the verification label gave students a reason to engage with AI-generated material without ending their scrutiny of it. Evaluations of AI learning resources that report only mean effects cannot detect whether the students the resources are meant to help are the ones who gain.
Read the full abstract

Artificial Intelligence

Small Language Models for Smart Data Model Classification at the Edge: A Cost-Aware Hybrid Approach

Cristian Martella, Angelo Martella, Antonella Longo et al. (+1 more)

The rapid proliferation of heterogeneous data sources within the Internet of Things (IoT) across domains such as smart cities, energy management, and environmental monitoring necessitates efficient and scalable data standardization methods. Effective classification of smart data models (SDMs) is essential for facilitating interoperability. However, existing approaches are often limited by high resource consumption and lack applicability in edge environments with constrained computational capabilities. Aiming to bridge this gap, the proposed study evaluates the performance of lightweight open-source language models (LMs) to resolve an input data entity against its corresponding best fitting SDM representation under resource-constrained conditions. It systematically benchmarks a diverse array of models, including general purpose (GP), reasoning-specialized (RS), and code-specialized (CS) architectures, across multiple domain-specific datasets. Addressing the current omission of lightweight, resource-efficient solutions in the literature, the investigation provides significant and valuable insights into model selection, task formulation, and deployment strategies that optimize accuracy and efficiency. A complementary experiment also compares the surveyed large language models (LLMs) against two near-zero-cost similarity baselines (Term Frequency-Inverse Document Frequency (TF-IDF) and a lightweight sentence encoder) on the same task, providing a strong reference point for interpreting the practical value of LLM-based classification on edge platforms.
Read the full abstract

All authors Cristian Martella, Angelo Martella, Antonella Longo, Motaz Saad

Artificial Intelligence

Smart Content Ingestion for Generative AI Workloads

Abbas Raza Ali, Muhammad Ajmal Siddiqui and Moona Zahid

The evolution of machine learning has progressively changed where intelligence resides in an AI system. In conventional machine learning the task, data representation, labels and model architecture were tightly coupled, so data preparation was narrow, schema-bound and visible. Generative AI decouples the model from any single task: one foundation model serves open-ended downstream tasks, and the generality gained on the model side is matched by heterogeneity on the data side, because enterprise knowledge is authored in the formats people use (PDF, presentations, spreadsheets, scanned documents, forms, tables, diagrams and mixed-layout files) that carry textual, visual, geometric and structural information at once. A language model or retriever cannot reason reliably over information misrepresented at this interface, so content extraction becomes a lifecycle stage in its own right whose errors no downstream retriever or re-ranker can repair. This paper presents a production-ready content-extraction system that makes this stage explicit, configurable, and measurable. The system incorporates selective OCR routing, a scarcity-first curation engine with a reference-based extraction scorer that measures character, word, and table-structure accuracy, a deterministic structure-aware parent-child chunker, and a read-only retrieval evaluator that generates grounded questions from every page and reports Hit@k, mean reciprocal rank, and latency. On a 180-document corpus the best extractor scores 97.4 of 100 (character error rate 0.13%, table similarity 0.995) and the chunker reaches hit@1 of 68.6%, hit@10 of 92.8% and MRR 0.77 over 25,050 generated questions. We distil three design principles (structure before semantics, never mutate what you measure, budget your labels) and position measured content extraction as the perception layer of enterprise agentic systems.
Read the full abstract

Artificial Intelligence · Cross-listed from cs.CR

Towards a Unified Misuse Monitoring Benchmark

Aniruddh Pramod, James Oldfield and Adel Bibi

LLM agents increasingly act in multi-actor environments, exposing them to misuse from multiple sources: decomposition attacks, where a harmful request is split into innocuous sub-requests, and prompt injection attacks, where a compromised tool delivers a malicious instruction. Existing evaluations treat these threats separately and ask whether a trajectory is harmful, rather than when it becomes harmful. We propose monitoring the agent's responses, where its actions are externalised, and ask whether the first point where monitors identify harm lands within a harm window (from the agent's first harmful commitment to goal execution). We develop a unified formalism for trace-level misuse monitoring and use it to construct a benchmark of ~6,200 conversation transcripts between a user, an LLM agent, and the external environment, spanning both threats in a shared schema, with a labelled harm window, corresponding benign controls, and matched instances of refusals to these requests. Across 17 monitor configurations, we find that our proposed action-framed monitors perform well on both threats under classical metrics (AUC: 0.95 and 0.99 respectively), while content-framed monitors collapse on injection attacks (AUC: 0.52). We also show that classical position-blind metrics paint an optimistic picture of monitor performance, since all monitors localise decomposition attacks poorly under the interval metric, which measures the ability to localise harm. Broadly, we illustrate the need for a unified study of misuse monitoring.
Read the full abstract

Artificial Intelligence · Cross-listed from cs.LG

SchemaFill: Efficient LLM Tool Calling via Slot-Parallel Speculative Decoding

Zhi-Kai Chen, Song-Yan Li, De-Chuan Zhan et al. (+1 more)

LLM agents interact with external systems by generating structured tool calls. Given a user request, conversational context, and a catalog of tool schemas, a tool-calling model must select tools and generate their arguments, potentially producing multiple calls in a single response. Standard autoregressive decoding generates these calls token by token, incurring substantial latency for requests involving multiple calls or many argument fields. The explicit argument structure offers opportunities for parallel generation, but later argument values may depend on preceding fields and calls, so independently generated values can differ from the target model's output. We present SchemaFill, a framework for efficient LLM tool calling through slot-parallel speculative decoding. SchemaFill generates future slot values concurrently as candidates, without requiring advance knowledge of the actual call sequence or argument values. Candidates spanning multiple fields and calls are concatenated for verification by the target model under the actual output prefix. Only verified tokens are committed, and the target supplies corrections when candidates disagree. This applies target verification while exploiting parallelism across slots and calls. On Glaive and BFCL, SchemaFill achieves up to a 4.05$\times$ improvement in end-to-end throughput over autoregressive decoding. Code is available at https://github.com/Czzzk/SchemaFill.
Read the full abstract

All authors Zhi-Kai Chen, Song-Yan Li, De-Chuan Zhan, Han-Jia Ye

Artificial Intelligence · Cross-listed from cs.CV

Graph-Based Recognition of Simulated Train-Driver States From Facial and Upper-Body Keypoints

Olivia Nocentini, Marta Lagomarsino, Gokhan Solak et al. (+6 more)

Driver fatigue poses a significant challenge to railway safety, with traditional systems like the dead-man switch offering limited and basic alertness checks. This study presents a vision-based monitoring system that relies solely on a single front-facing RGB camera and a graph neural network to classify simulated train-driver states into alert, not-alert, and an emergency class comprising acted emergency-like behaviours. To optimize input representations for the model, an ablation study was performed, comparing three feature configurations: skeletal-only, facial-only, and a combination of both. Experimental results show that combining facial and skeletal features yields the highest accuracy (81%) for the three-class model under the light condition, outperforming models that use only facial or skeletal features. Furthermore, the combination of facial and skeletal features achieves 99% accuracy in the alert/not alert classification in light condition. Additionally, we introduced a controlled RGB video dataset containing alert, not alert, and acted emergency-like behaviours recorded under three illumination conditions. These contributions represent a step toward passive and non-contact train-driver state recognition based on facial and upper-body dynamics.
Read the full abstract

All authors Olivia Nocentini, Marta Lagomarsino, Gokhan Solak, Younggeol Cho, Qiyi Tong, Sara Zeynalpour, Marta Lorenzini, Alessandro Ledda, Arash Ajoudani

Artificial Intelligence · Cross-listed from cs.RO

Demo: Vision-Language Model-Guided Online Calibration of an Electromagnetic Digital Twin

Zerui Kang, Yishen Lim, Zhouyou Gu et al. (+4 more)

An electromagnetic (EM) digital twin gives mobile robots wireless situational awareness but depends on material conductivities that change with the environment. Online calibration faces initialization sensitivity and measurement travel costs. We demonstrate a vision-language model (VLM)-guided framework using a Unitree G1 robot and NVIDIA Sionna, with two VLM calls: material classification maps visible materials through ITU-R P.2040 to conductivity priors for Sionna's gradient descent on accumulated received signal strength (RSS) measurements; waypoint planning selects the next measurement location online using residual RSS calibration error and image coverage. In a real indoor scenario, the framework achieves a normalized mean absolute conductivity error of $1.74\times10^{-4}$ within 20 m of travel; random initialization never converges, while random waypoints require over twice the travel.
Read the full abstract

All authors Zerui Kang, Yishen Lim, Zhouyou Gu, Seungnyun Kim, Seung-Woo Ko, Tony Q.S. Quek, Jihong Park

Artificial Intelligence

CuratorMAS: Automating Dataset Curation via Multi-Agent Orchestration

Yixin Zhang and Wenjie Feng

High-quality datasets are essential for reliable machine learning, but dataset curation remains costly and hard to generalize across domains. Existing methods typically rely on manually designed heuristics or model-dependent signals, limiting their applicability across tasks and user queries. To address these limitations and automate data curation, we propose CuratorMAS, a multi-agent collaboration framework that orchestrates agents to evaluate and curate high-quality datasets. To achieve the goal of flexible curation, CuratorMAS decomposes the complex curation process into five programmable execution stages and forms a parallelizable workflow. Specifically, CuratorMAS first performs dataset exploration to collect contextual information such as file structures and constraint cues, thereby developing a comprehensive understanding of the given task. In order to acquire up-to-date information, CuratorMAS retrieves domain knowledge from online sources to augment the evaluation process. Next, CuratorMAS derives the necessary evaluation criteria and computes the corresponding metrics. Based on these results, CuratorMAS executes filtering accordingly. Finally, an evolution module summarizes the evaluation outcomes and updates the relevant skills. Extensive and comprehensive experiments demonstrate that CuratorMAS significantly reduces the noise rate by up to 36.03 percentage points (pp) while also improving the F1 score of downstream models by up to 8.88 pp.
Read the full abstract

Artificial Intelligence · Cross-listed from cs.LG

Learning to Simulate Individuals from Macro Social Signals

Yining Zhao, Bushi Liu, Haofei Yu et al. (+5 more)

Large language models are increasingly used to simulate how individuals respond to new situations, yet the behavioral reasoning behind these responses is either inherited from pretraining or learned from individual-level annotations, which offer limited behavioral diversity and little supervision of the reasoning itself. We propose to learn behavioral reasoning from prediction markets, whose price trajectories record how populations respond to real-world events at scale. We introduce macro2mind, which trains a language model with GRPO using market signals. A social behavioral decomposition makes behavioral reasoning an explicit step of forecasting: the model infers representative groups of market participants, predicts how each interprets the news and updates its beliefs, reasons about their interactions, and aggregates these responses into a price. A hindsight-regret curriculum with difficulty-aware sampling focuses training on transitions where hindsight-identified groups substantially improve the forecast while prioritizing examples that remain learnable for the current policy. The learned reasoning applies to user simulation without further training. On SWM-Bench, macro2mind achieves state-of-the-art directional accuracy and correlation on Polymarket. Trained on market data, it transfers zero-shot to four user-simulation benchmarks (Humanual, OvertonBench, PRISM, and CAD) and has competitive performance among zero-shot methods. Used as a data generator, macro2mind also raises a downstream simulator's accuracy on unseen users by 15.5 points, outperforming data generated by its backbone by 13.2 points.
Read the full abstract

All authors Yining Zhao, Bushi Liu, Haofei Yu, Zhengyang Qi, Shanyong Wang, Chuyue Li, Yuxiang Liu, Jiaxuan You

Artificial Intelligence · Cross-listed from cs.LG

TRIAGE: Direction-Aware Mismatch Stabilization of Native NVFP4 Reinforcement Learning

Zhen Li, Shuai Zhang, Yanggan Gu et al. (+7 more)

Low-precision execution can substantially accelerate reinforcement learning (RL) for large language models, but discrepancies between learner and sampler execution can destabilize policy optimization. In this paper, we characterize the interaction between mismatch and the policy-gradient direction, distinguishing locally amplifying from contracting update contributions that mismatch magnitude alone cannot identify. In native NVFP4 runs, we observe an early imbalance between the two amplifying regions, favoring negative-advantage, negative-gap updates. Their tail tokens become concentrated in a small fraction of response segments before mismatch spreads globally. Motivated by these findings, we introduce TRIAGE, a direction-aware stabilization method that uses segment-level diagnosis to selectively rebalance policy-gradient updates and applies bounded repair to residual severe mismatch. TRIAGE modifies the optimization objective while retaining native NVFP4 weight-and activation 4-bit (W4A4) forward execution on both the sampler and learner. Experiments on Qwen3-4B and Qwen3-30B-A3B show stable optimization throughout the evaluated training horizon and achieve full precision level performance across five mathematical reasoning benchmarks, while native NVFP4 with TRIAGE provides up to 2.3x higher rollout throughput than BF16.
Read the full abstract

All authors Zhen Li, Shuai Zhang, Yanggan Gu, Yiming Zhang, Yang Yu, Mingfa Feng, Congkai Xie, Shuang Yu, Junjie Lai, Hongxia Yang

Artificial Intelligence · Cross-listed from cs.CV

One Figure, Every Canvas: Editable Flowchart Relayout via Agentic Pipeline

Shih-Chen Tseng, Chih-Hsuan Chen, Ryan Yang et al. (+3 more)

Pipeline figures in ML papers must be repurposed across many canvases, including paper columns, 16:9 slides, portrait posters, 1:1 social teasers, 9:16 phone previews. Each format imposes a different aspect ratio on the same computational graph, where any silently broken connection misrepresents the method. We formulate aspect-ratio-adaptive flowchart relayout as a distinct task: given a raster flowchart and a target ratio, produce a structurally faithful, hallucination-free, editable layout. Existing methods fail characteristically: image-to-image models stretch blocks and reject extreme ratios, text-to-image agentic systems hallucinate content, and parse-then-render systems mis-route edges. We propose an agentic pipeline factored into Parse, Style, and Layout stages, each pairing a main agent with a critic that combines deterministic constraint checks with VLM visual feedback so connectivity is explicitly checked and prevented from being silently broken. Outputs are draw.io-editable mxGraph XML. On a curated benchmark of 100 flowcharts at five aspect ratios, evaluated by Gemini 3.1 Pro and validated against human judgments, our method reaches 68.6% Content Fidelity versus 11.2-41.4% for prior work. Project page: https://onefigureeverycanvas.vercel.app/
Read the full abstract

All authors Shih-Chen Tseng, Chih-Hsuan Chen, Ryan Yang, Hsi-An Chen, Chun-Wei Tuan Mu, Yu-Lun Liu

Artificial Intelligence · Cross-listed from cs.LG

Base Models Can Reason By Taking a Cue From Training Data

Sophie L. Wang, Amil Dravid, Rulin Shao et al. (+3 more)

In this paper, we study how training data creates associations between the tokens at the start of a base model's response and the reasoning behavior that follows. First, we demonstrate that fixing particular starting token cues makes a base model's performance competitive with that of its reinforcement learning (RL)-trained counterparts on math and coding. For instance, the cue ".\n\nOkay" raises Olmo-3-7B's MATH-500 pass@1 accuracy from 42% to 78%, while "Alright," raises Qwen3-14B's from 72% to 87%. Second, RL makes these cues more likely, while fixing them recovers much of its performance gain over the base model. Third, we trace the reasoning effects of token cues to the training data. We perform causal data interventions to turn an arbitrary word, such as "chicken", into an effective reasoning cue, or remove an existing cue's effect. A similar edit makes the prompt instruction "Think duck duck goose" as effective as "Think step by step" at eliciting reasoning. We also find that the hidden state representations induced by different cues correlate with different document types from the training set. Finally, we extend our study of token cues with a case study in language model safety, finding that different cues elicit distinct refusal and compliance behaviors that correspond to different types of training data.
Read the full abstract

All authors Sophie L. Wang, Amil Dravid, Rulin Shao, Kevin Farhat, Sewon Min, Alexei A. Efros

Artificial Intelligence

BiasFlow: Geometric Monitoring and Backbone Regularization for Spurious Feature Reliance

Haojin Deng, Zhiping Lin and Yimin Yang

Worst-group accuracy (WGA) evaluates a trained predictor but does not characterize how its frozen backbone behaves when a new head is learned. We introduce BiasFlow, a hook-based toolkit for monitoring class-attribute centroid alignment (IBMI), within-class centroid separation (W-IBMI), and feature-projection sensitivity. IBMI is confounded by class-attribute correlation and is not a measure of causal feature reliance. We pair these diagnostics with BiasFlow Regularization (BFR), a supervised, composable class-conditional centroid-alignment penalty. W-IBMI verifies the quantity BFR optimizes; it is scale dependent and does not independently establish attribute removal. Across the reported small-scale benchmarks, adding BFR improves or preserves mean WGA, with gains up to +26.0 pp on UrbanCars. The principal independent stress test freezes CelebA-Std backbones and trains fresh heads on biased data: BFR+GroupDRO improves WGA from 40.7% to 64.1%, while Male probe accuracy decreases from 92.5% to 72.2%. Attribute information remains recoverable, and cross-task results are mixed. A controlled synthetic-watermark ImageNet experiment additionally improves watermark-shift accuracy by +23.0 pp under matched training. These results support evaluating centroid geometry and resistance to biased head retraining alongside WGA, within the tested protocols.
Read the full abstract

Artificial Intelligence · Cross-listed from cs.CV

Learning to Read the Contextual Tokens in Diffusion Transformers

Omer Dahary, Etai Sella, Hadar Averbuch-Elor et al. (+2 more)

Multimodal Diffusion Transformers (MM-DiTs) jointly process visual and textual representations throughout generation. These models repeatedly update the text tokens through multimodal attention, forming dynamic contextual tokens whose function is not well understood. In this work, we introduce a framework for reading this contextual space through natural-language interrogation. We train a lightweight bottleneck network that maps intermediate contextual tokens into the input space of a frozen Large Language Model (LLM), allowing the LLM to answer questions about the emerging image directly from these hidden representations. Our reader reveals that contextual tokens encode a rich, global representation of the emerging scene: generation-specific semantics, including attributes left underspecified by the prompt, are accessible surprisingly early in denoising, while increasingly fine-grained details become readable over time. Remarkably, this information remains decodable even when the MM-DiT receives an empty prompt, showing that contextual tokens accumulate substantial image-specific information from the evolving visual representation itself. We further find that generations with more readable contextual representations tend to receive higher human-preference scores. Building on these observations, we introduce Contextual Alignment, a training technique that explicitly reinforces the visual-semantic information encoded in the contextual tokens, improving generation quality and distributional coverage. Together, our results establish contextual tokens as both an interpretable view into the internal dynamics of MM-DiTs and an effective target for improving generative models.
Read the full abstract

All authors Omer Dahary, Etai Sella, Hadar Averbuch-Elor, Daniel Cohen-Or, Or Patashnik

Artificial Intelligence · Cross-listed from cs.RO

Recursive Video In-Context Learning for Agentic Robot

Wenrui Bao, Xinxin Liu, Bingxin Xu et al. (+1 more)

LLM agents that orchestrate frozen vision-language-action (VLA) policies improve across episodes through text memory, which records what the agent did but not how the task is done. A demonstration video shows it, but fits poorly into an agent's context. The full video slows every turn, fixed keyframes lose the contact detail that decides whether a grasp holds, and what the agent needs shifts from the task's structure while planning to the frames around each contact. We introduce Recursive Video In-Context Learning (RV-ICL), a training-free method that turns a demonstration into a hierarchy the agent navigates rather than a prompt it receives. The hierarchy is built from the sub-events of the demonstration, such as grasps and releases. Its levels grow finer, from keyframes of the whole task to phases, moments and short clips, and are exposed through read-only tools. The agent reads the coarse levels before planning. During execution it re-enters the hierarchy whenever a step needs more detail and loads only the clip of its current sub-goal. One demonstration per task is enough. Built on RPent, RV-ICL raises success from 92.6% to 96.5% on LIBERO-PRO and from 86.7% to 95.8% on LIBERO-Plus.
Read the full abstract

All authors Wenrui Bao, Xinxin Liu, Bingxin Xu, Yuzhang Shang

Artificial Intelligence · Cross-listed from cs.CV

UniSlider: Perceptually Uniform Sliders for Continuous Image Editing

David Serrano-Lozano, Duygu Ceylan, Yannick Hold-Geoffroy et al. (+3 more)

Sliders provide an intuitive interface for continuous image editing. In current generative approaches, however, the slider is simply a rescaling of the method's strength parameter, such as an adapter coefficient, a prompt weight, or an interpolation factor. This strength relates poorly to perceptual change. The image can partially revert as the slider moves, long stretches of the range produce no visible difference, and short intervals transform the image abruptly. Remapping the strength could fix this uneven pace, but only if the trajectory is monotone, which current methods do not enforce. We therefore distinguish the slider from the strength, and require perceptual distance from the input to grow linearly with the slider value. We introduce UniSlider, a lightweight LoRA trained on a few-step editing backbone so that its strength approximates this ideal slider. Few-step sampling lets us impose this objective in pixel space without intermediate ground truth, and the backbone's output is preserved at full strength. However, a low-rank adapter cannot make the strength fully uniform. Our slider is thus an inference-time remapping of the strength, obtained by adaptive sampling. Since training optmizes to make the trajectory monotone, this remapping closes the remaining gap without extra training or parameters. On a new benchmark of 300 continuous edits evaluating uniformity, monotonicity, edit fidelity, and identity preservation, UniSlider outperforms all prior methods and is preferred in a user study.
Read the full abstract

All authors David Serrano-Lozano, Duygu Ceylan, Yannick Hold-Geoffroy, Iliyan Georgiev, Javier Vazquez-Corral, Anna Frühstück

Artificial Intelligence · Cross-listed from cs.CL

MemPilot: Orchestrating On-Demand Multimodal Memory Curation for LLM Agents

Haozhen Zhang, Haodong Yue, Quanyu Long et al. (+6 more)

Memory has become integral to the LLM agent ecosystem, supporting information retention and reuse across interactions. However, most existing agent memory systems construct memory in a query-agnostic manner, which can incur unnecessary preprocessing cost and discard details that later prove essential. Recent studies have begun shifting memory processing toward runtime adaptation, but typically specialize in particular operations or fixed processing schemes, leaving flexible control over performance, cost, and latency largely underexplored. To address this challenge, we present MemPilot, a flexible framework that orchestrates on-demand memory curation under different performance–cost–latency preferences. Specifically, we optimize a multi-step LLM policy via reinforcement learning to iteratively choose between retrieving from query-agnostic memory and delegating query-specific curation of raw multimodal history to heterogeneous LLMs and VLMs. The policy jointly controls evidence amount, curation instructions, model selection, and visual access, enabling fine-grained allocation of runtime computation. To optimize this policy under competing objectives, we adapt objective-wise advantage decoupling by separately estimating each objective's advantage before aggregation. Moreover, we introduce prefix-based marginal utility estimation for fine-grained credit assignment across multi-step rollouts. Experiments on five multimodal agent-memory benchmarks demonstrate favorable performance–cost–latency trade-offs across optimization preferences, with preference sweeps yielding broader frontiers than existing trade-off-aware baselines.
Read the full abstract

All authors Haozhen Zhang, Haodong Yue, Quanyu Long, Jianzhu Bao, Qingyuan Liu, Tao Feng, Bohan Liu, Weida Liang, Wenya Wang

Artificial Intelligence · Cross-listed from cs.CL

CLIFT: Conformal Self-Verification for Web Agent Training and Test-Time Scaling

Yifan Zhang, Yutong Dai, Viraj Prabhu et al. (+3 more)

Open-source web agents are now strong enough to execute realistic browser tasks, but training them with reinforcement learning still depends on weak supervision: binary task success is too sparse for credit assignment, while frontier-language-model judges are too expensive to call at every step and cannot be assumed available at deployment. We introduce CLIFT, a training and test-time scaling method built around conformal self-verification. During training, the agent answers natural-language verification questions about its own rollouts; a Compositional Conformal Certifier keeps only question signals whose URL-conditional evidence agrees with a training-time judge, assigns signed trust weights through polarity-aware lift, and blends the resulting verifier score into per-step rewards in a way that never subtracts from the judge baseline. At test time, the same certified bank is frozen and reused as structured evidence for Conformal Trajectory Selection (CTS): the agent samples a greedy rollout and one or more diverse retries, the self-verifier summarises each URL trace, and a conservative majority-vote rule chooses whether to swap away from the current incumbent without calling any external judge. This single mechanism supports three settings. On WebArena Infinity, CLIFT achieves state-of-the-art performance among open-source web agents. On VisualWebArena, a bank trained with the open model transfers to GPT-5.5 at test time and reaches state-of-the-art performance under the canonical harness. On Online Mind2Web, without training an agent on the benchmark, translating the certified question bank improves a live-web agent in zero-shot evaluation. Together these results position conformal self-verification as a way to turn costly judge feedback into a reusable training signal and a judge-free test-time scaling signal.
Read the full abstract

All authors Yifan Zhang, Yutong Dai, Viraj Prabhu, Zhiyuan Hu, Ran Xu, Zeyuan Chen

Artificial Intelligence

TasteVal: Measuring the Experimental Research Taste of AI Systems Against Human Experts

Oliver Jaffe and Dane Sherburn

We introduce TasteVal, a benchmark to evaluate the experimental research taste of frontier models. We define research taste as the ability to pick interesting problems to solve, design experiments, and interpret experimental results. TasteVal measures the experimental component of research taste; given a fixed research problem, we measure how well a model iteratively designs experiments and draws conclusions from their outcomes. We operationalize experimental research taste as compute efficiency; a Researcher who reaches the same score as an expert human using half the serial experimental compute has twice the experimental taste. Experimental taste thus acts as a multiplier on experimental compute, making it a key input to forecasts of AI progress. TasteVal consists of 8 novel, challenging, open-ended tasks representative of frontier AI R&D. To isolate taste from coding ability, the model under evaluation acts as a Researcher that iteratively designs experiments while a fixed Coder agent implements them and reports their results. The Researcher executes until either the 40 H100 hour or 120 wall-clock hour budgets are exhausted. We recruit 24 human experts, at least 2 per task, and take the best expert attempt per task as the expert baseline. We evaluate 20 models released between 2023 and 2026. The best-performing model, Opus 5.5, exceeds our expert baseline, with a compute multiplier of 2.3x (95% CI 1.15-4.37), at roughly 1/30 of our baseliners' average per-run cost. On TasteVal, the compute multiplier of frontier models has doubled approximately every 3.0 months since December 2025 (95% CI 1.7-5.0), up from every 14 months between 2023 and December 2025. Measured by final normalized performance, frontier models show no trend break, doubling every 14.6 months. To keep TasteVal uncontaminated, we do not release the tasks.
Read the full abstract

Artificial Intelligence · Cross-listed from cs.LG

Deep Learning for Sleep Heart Rate Estimation from Accelerometers: Toward Population-Scale Cardiac Insight Without Optical Sensors

Tanbin Islam Rohan, Pranjol Sen Gupta, Tanusree Debi et al. (+1 more)

Large longitudinal cohorts often contain wrist accelerometry without optical heart-rate sensing, motivating recovery of cardiac information from motion signals already collected during sleep. We present SeqSmoother, a transformer-based temporal corrector for sleep heart rate (HR) estimation from wrist accelerometry. SeqSmoother combines spectral descriptors with an intermediate Nightbeat-derived frequency anchor and a physics-motivated sub-harmonic feature designed to identify harmonic frequency lock-on. All inference-time features are derived from wrist accelerometry, while ECG is used only to construct reference HR labels and training-label quality weights. We evaluate SeqSmoother using 13 participant-disjoint held-out folds and compare it with the official Nightbeat implementation under a matched 60-s window and 15-s step protocol. Across all out-of-fold predictions, SeqSmoother achieved a participant-macro MAE of 1.60 bpm. On Nightbeat-retained matched intervals, Nightbeat achieved lower absolute error than SeqSmoother (0.615 versus 1.091 bpm), while SeqSmoother provided estimates over a larger portion of the eligible recording; Nightbeat produced final estimates for 72.85% of the SeqSmoother-eligible out-of-fold grid. Separately, the proposed sub-harmonic ratio achieved an AUROC of 0.972 for identifying reference-defined harmonic lock-on candidates. These findings reveal an accuracy-availability trade-off between learned temporal modeling and quality-gated signal processing while providing empirical support for a physics-informed approach to identifying frequency-tracking failures in accelerometer-based sleep HR estimation.
Read the full abstract

All authors Tanbin Islam Rohan, Pranjol Sen Gupta, Tanusree Debi, Nazmus Sakib

Artificial Intelligence · Cross-listed from cs.SD

Paradee: Distilling Kokoro-82M into an 8M-Parameter Single-Voice Text-to-Speech Model

Sahil Mahendrakar

We distill Kokoro-82M, a widely used open text-to-speech model with 54 voices, into Paradee, an 8.07M-parameter model that speaks one of them. Paradee keeps Kokoro's architecture with much narrower layers, and each of its two halves is trained separately against the frozen teacher. It has 10x fewer parameters and needs 15x less compute. We first synthesize a corpus with the teacher and keep its durations, pitch, energy and phoneme features. We then train a small text side to predict these values, and a small decoder to turn the teacher's saved values into the teacher's audio, first with spectral losses and then adversarially. Finally, we connect the two halves and quantize the weights to int8. It needs no alignment learning and no joint training, and it runs on one laptop. Stored in int8, Paradee is 8.5 MB, runs 25x faster than real time on one CPU thread, and scores 4.41 on UTMOS against the teacher's 4.52. The student initially kept a slight buzz, which we trace to the phase of voiced speech between 2 and 8 kHz. A phase-locking filter applied after synthesis removes most of it, with no training and no extra parameters. Code, model files and audio samples are at https://github.com/sahilmahendrakar/paradee
Read the full abstract

Artificial Intelligence · Cross-listed from cs.CV

TAPDreamer: Transferable Adversarial Patches for World Action Models

Xuanyu Lu, Fengqing Jiang, Kaiyuan Zheng et al. (+8 more)

World models learn to predict how their environment will evolve, making them an important foundation for general-purpose robotic control. Yet world action models depend on camera inputs whose manipulation can corrupt the visual representations used across tasks and action policies. Existing attacks on these models optimize against the victim's actions or predicted futures and therefore require access to target-model outputs. In this paper, we propose an attack, TAPDreamer, against world action models that instead uses a public encoder alone to construct a fixed local perturbation that transfers across tasks and action architectures. TAPDreamer requires no target-policy queries. Our key insight is that interactions between patch-induced changes in attention weights and value vectors broadcast a nearly identical representation shift far beyond the patch footprint, and this shift remains stable across task observations. Guided by this insight, TAPDreamer uses six frames from one source task to maximize the global L1 distance between clean and patched encoder representations. In closed-loop evaluation, one frozen patch per benchmark, covering about 6.5% of the input, reduces FastWAM's success rate from 97.7% to 0.0% across 40 LIBERO tasks and from 90.86% to 0.0% across 50 RoboTwin tasks; matched random patches retain 81.5% and 79.2% success. The same patches reduce success to 1.45% and 1.00% on two DreamWAM configurations and to 10.60% on Motus. These results show that protecting downstream action generation alone is insufficient: defenses for world action models must also secure shared visual encoders against persistent local perturbations.
Read the full abstract

All authors Xuanyu Lu, Fengqing Jiang, Kaiyuan Zheng, Yichen Feng, Yaorui Ding, Yuetai Li, Zhen Xiang, Bhaskar Ramasubramanian, Basel Alomair, Luyao Niu, Radha Poovendran

Artificial Intelligence · Cross-listed from cs.LG

Sharpen Without Search: On-Policy Distillation of Sequence-Level Power Distribution

Erfan Baghaei Potraghloo, Seyedarmin Azizi, Arya Fayyazi et al. (+4 more)

A language model can give a correct answer more probability than any single incorrect answer and still usually sample an incorrect one, because the incorrect answers together hold more probability. The power distribution raises each complete answer's probability to a power above one and renormalizes, shifting probability toward answers the model finds most likely (sharpening). Sampling from it improves reasoning without changing parameters, but needs many scored candidates per query. We show that a model can instead be trained to produce such answers in one generation. On-policy power distillation (OPPD) runs a sequential Monte Carlo sampler in which the model being trained generates candidates and a frozen teacher's power distribution weights them; the same probabilities weight each answer in a maximum-likelihood update. Training raises single-generation accuracy by up to 23.0 points on MATH500 and 27.3 on GSM8K over the untrained model at the same temperature, and one generation scores 2.4 and 3.5 points above published power sampling with 64 candidates, recovering 94 percent of the gain that 16 candidates give the untrained model. For context, against GRPO trained with verified rewards from the same checkpoint and budget, OPPD scores 3.8, 4.0 and 5.4 points higher on MATH500, GSM8K and AIME using no reference answers; the two are complementary, and OPPD applied after GRPO adds up to 9.3 points. Trained only on mathematics, OPPD raises HumanEval accuracy by up to 5.3 points. One loss coefficient moves the sharpening exponent the model absorbs between 1.19 and 2.02, against 1.14 for ordinary on-policy distillation, and it rises mostly on the model's own answers. Gains hold across model families and sizes, including a model already trained with verified rewards, where lowering the temperature gives nothing and OPPD adds 4.4 points on MATH500. Code: https://github.com/ArminAzizi98/OPPD.
Read the full abstract

All authors Erfan Baghaei Potraghloo, Seyedarmin Azizi, Arya Fayyazi, Saeid Shokoufa, Mehdi Kamal, Souvik Kundu, Massoud Pedram