Computation and Language · Cross-listed from eess.AS
GIVE-KWS: Gated Injection of Visual Evidence for Noise-Robust Query-by-Example Keyword Spotting
Ming-Hsiang Hu, Kuan-Tang Huang, Hung-Shin Lee et al. (+1 more)
Visual
speech promises noise-robust keyword spotting, yet a visual stream is not necessarily used. On a tri-modal query-by-example keyword spotting (QbyE-KWS)
benchmark, we find that a system with a task-trained visual encoder comes within 2 percentage points of a text-and-audio system in equal error rate (EER) at -10 dB, and link this gap to the encoder's lack of phonemic information. We present GIVE-KWS, whose fusion stage, GIVE (Gated Injection of Visual Evidence), conditions query audio on lip motion through gated cross-
attention. We show that visual robustness depends on two interacting conditions: a phoneme-bearing visual representation, and fusion that injects visual evidence rather than rescaling audio features. Under a phoneme-bearing encoder, injection yields an effective SNR gain of 4.0-9.3 dB over masking at -10 dB, whereas under a phoneme-poor one it nearly vanishes. Relative to the
benchmark system, GIVE-KWS reduces unseen-keyword EER by 72.9% at -10 dB and 62.8% on average.
Read the full abstract ▾
All authors Ming-Hsiang Hu, Kuan-Tang Huang, Hung-Shin Lee, Berlin Chen
Formal Languages and Automata
The compress-with-another threshold of Szykuła's Figure 3 family
Qichao Wang
We give a self-contained pair-automaton proof of the exact compress-with-another threshold of the corrected Figure 3 family from a recent survey of open problems in synchronizing automata. For every $p \geq 3$, the automaton has $n = 3p$ states and $\mu(q_0) = 4p = 4n/3$. The word $(ba)^p(ab)^p$ attains this value. Two entrance potentials and an excluded region yield the lower bound, with the endpoint exception at $p = 3$ treated explicitly. A separate reset construction proves $\operatorname{rt}(A_p) \leq 3p^2 + 4p - 1$ for every parameter. Reproducible computations verify the transition and entrance identities; the all-parameter results follow from the explicit proofs.
Read the full abstract ▾
Machine Learning
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
Human-Computer Interaction
Tellimation: Making Narrative Gaps Visible in Children's Storytelling with Just-in-Time Animation
Vincent Cavez, Marielle Zheng, Momin Siddiqui et al. (+4 more)
Illustrated scenes are used to
prompt children's stories, yet children with
language difficulties often omit the details, actions, and relationships that make a story coherent. We introduce Tellimation, which animates scene elements a child has omitted or misdescribed, drawing
attention to a narrative gap without saying what to tell. Its design space covers eight kinds of gaps (who is in the scene, how many, what they are like, what they are doing, where, when, how they relate, and what lies beyond the picture, such as
speech and thoughts), instantiated through 20 parameterized animations grounded in classical animation principles. A real-time pipeline detects discrepancies between utterance and scene, then selects and parameterizes an animation from the scene's narrative potential and the child's history. Adults interpreted most animations without instruction (N=120); children (N=12) resolved 46% of the gaps the system identified with animations, against 13% without.
Read the full abstract ▾
All authors Vincent Cavez, Marielle Zheng, Momin Siddiqui, Abbie L Olszewski, Srirangaraj Setlur, Maneesh Agrawala, Hari Subramonyam
Computer Vision
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
Machine Learning
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
Robotics
InterMimicGen: Scaling Humanoid Loco-Manipulation through Self-Evolving Motion Imitation
Yucheng Zhang, Sirui Xu, Jinhong Li et al. (+8 more)
Captured human-object interactions provide rich supervision for
humanoid loco-manipulation, but they are sparse, heterogeneous, and not directly executable by
robots. We introduce InterMimicGen, a self-evolving motion-imitation framework in which
robot motion data and a tracking policy improve each other. First, we consolidate motion-captured human-object interaction
datasets and retarget them into
humanoid robot references while preserving whole-body coordination and dexterous hand-object relationships. This produces a large and diverse
humanoid robot reference collection for dexterous whole-body loco-manipulation. Second, we train a physics-based generalist tracker that executes these references in simulation on a
humanoid with dexterous hands, covering a scale and diversity beyond prior
humanoid tracking systems for loco-manipulation. Third, we close a data flywheel: each round makes small, task-preserving changes to where an interaction takes place and how the body performs it, fine-tunes the tracker on them, and keeps only the variants whose simulated execution completes the task, which seed the next round. With more iterations, these small edits compound into broader coverage around the sparse original demonstrations while preserving task semantics and motion quality. Experiments show contact-preserving retargeting across
robot configurations, broad tracking with a single generalist policy, executable motions that keep growing over augmentation rounds, and transfer to real
robots. InterMimicGen provides a unified path from heterogeneous human demonstrations to a continually expanding motion resource for
humanoid robot learning.
Read the full abstract ▾
All authors Yucheng Zhang, Sirui Xu, Jinhong Li, Liuyu Bian, Anatulya Nandi, Derek Zhang, Xiangchen Liu, Xueting Li, Umar Iqbal, Yu-Xiong Wang, Liang-Yan Gui
Cryptography and Security
TranScope: What the Software Hides About LLM Training Data, the Hardware Reveals at Scale, and Accelerators Magnify
Joshua Kalyanapu, Darsh Asher, Kaushal Mhapsekar et al. (+5 more)
Membership is the root
privacy primitive in machine learning: to date, no hardware-based out-of-distribution detection on black-box models has been demonstrated against constant-time, static neural networks with masked confidence. This paper performs the first cycle-level examination of how large
language models and
vision transformers interact with various modern microarchitecture components, including integrated accelerators, as LLMs scale in size and answers the question of whether the data that a model was trained on affects its execution footprint even without any input-dependent branch, dynamic optimization, or early exit and in constant-time models. The results confirm that the answer is yes and identify which modern hardware components, such as TLBs or on-core accelerators, reveal or amplify that effect. The results also answer whether the signal is informative enough to reliably classify the in-/vs/out-of-distribution property of membership. To understand why, we perform a systematic root cause analysis and find that the
transformer's tokenization steps, which happen during training, alter the locality of the accesses the model makes to fetch the vocabulary
token later during inference and, as a result, change the page table access patterns and TLB in a previously unknown data-dependent way, causing microarchitectural state to vary significantly based on whether or not the input was in the distribution of the
transformer training data. Building on the above observation, we introduce TranScope: the first microarchitecture tool for detecting membership information with low cost, no need for a surrogate model, and significantly higher robustness, e.g., 0.6 AUC for PETAL (best previously reported) vs 0.9 AUC (ours). This reintroduces hardware as both an opportunity, e.g., a tool for checking copyright violation for the first time, and a new channel for inferring membership (MIA).
Read the full abstract ▾
All authors Joshua Kalyanapu, Darsh Asher, Kaushal Mhapsekar, Bita Aslrousta, Rushiraj Chaitanyakumar Sheth, Maharshi Mukeshkumar Oza, Achyuta Kannan, Samira Mirbagher Ajorpaz
Computer Vision
S2PD: Serial-to-Parallel Diffusion for Physically and Logically Consistent Video Generation
Jeffrey Hu, Daniel Olmeda Reino and Ayush Tewari
Bidirectional
video diffusion models denoise entire
videos in parallel, yet when trained on effectively unlimited in-distribution data from procedural generators, continue to violate physical laws and simple symbolic rules. We introduce Serial-to-Parallel
Diffusion (S2PD), which performs autoregressive
diffusion at high noise before switching to parallel
diffusion at low noise. The autoregressive phase provides the serial computation needed to coordinate interdependent events and produce valid state transitions while the parallel phase jointly refines the entire
video and reduces sampling time relative to fully serial generation. We implement S2PD with two architectures: a pixel-space
diffusion transformer trained from scratch and a pretrained
video model adapted through LoRA fine-tuning with causal
attention. Across games, physical simulations, and real
video, S2PD follows rules more reliably than matched bidirectional baselines and generates
videos with greater temporal stability and sampling efficiency than other serial methods.
Read the full abstract ▾
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 ▾
Computer Vision
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
Robotics
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
Information Theory · Cross-listed from quant-ph
A Spectrum-Based Converse for Quantum State Discrimination and Its Applications to Classical-Quantum Channel Coding
Tamás Havas, Hsuan-Yin Lin and Eirik Rosnes
We investigate converse bounds on the average decoding error probability in finite-blocklength classical-
quantum channel coding. We first present a lower bound for multiple
quantum hypothesis testing in terms of pairwise trace distances and derive a corresponding fidelity bound. We then obtain a spectrum-based converse that depends only on the a priori probabilities and spectra of the states. We show that the converse bound remains tight for the
quantum depolarizing channel under suitable conditions. For codes with product-state outputs, this converse takes an explicit form involving products of output-state
eigenvalues. We apply it to binary codes over the
quantum amplitude damping channel using the input states $\vert+\rangle$ and $\vert-\rangle$. For this setting, we also discuss a normal approximation to the spectrum-based converse in the large blocklength regime. In all numerical examples considered, the spectrum-based converse is tighter than the other converse bounds at low noise levels.
Read the full abstract ▾
Computer Vision
Anatomy-aware Fine-grained Multimodal Fusion for Laryngopharyngeal Cancer T-Staging Prediction Using CT and Radiology Report
Xingyue Zhao, Yanzhou Su, Fang Zhang et al. (+12 more)
Accurate T-staging is crucial for guiding personalized treatment strategies for laryngopharyngeal cancer. However, current clinical practice relies on invasive biopsy procedures, whereas CT-based staging remains challenging due to the complex patterns of tumor invasion. Recent computer-aided approaches face two key challenges: 1) Structural relationship modeling: existing methods underrepresent anatomically structured patterns of tumor invasion, as they either process whole CT volumes without tumor-specific anatomical constraints or rely on labor-intensive tumor
segmentation. 2) Fine-grained cross-modal
alignment: while radiology reports contain organ-specific invasion details, current methods that apply global feature fusion struggle to accurately align individual anatomical structures with their corresponding textual descriptions. To address these issues, we propose an anatomy-aware
multimodal framework that integrates organ-level CT context and radiology reports into a unified representation for laryngopharyngeal T-staging. The framework first constructs an Anatomy-Structured Organ Graph (AOG) that captures invasion patterns between primary sites and surrounding organs, then performs Organ-Anchored Cross-Modal
Alignment (OCA) so that each organ node aggregates textual evidence from the radiology report, and finally refines this graph representation by injecting organ-specific invasion cues extracted from the report via Report-Enhanced Graph-Refinement (REG), yielding a
multimodal organ graph that combines spatial and textual evidence. Extensive experiments demonstrate that the proposed framework achieves superior performance in T-staging of laryngopharyngeal cancer.
Read the full abstract ▾
All authors Xingyue Zhao, Yanzhou Su, Fang Zhang, Zhanghexuan Ji, Yirui Wang, Dazhou Guo, Sibo Ju, Yuehua Cheng, Yuzhen Chen, Ming Feng, Le Lu, Tsung-Ying Ho, Jian Wang, Dakai Jin, Na Shen
Computational Complexity · Cross-listed from quant-ph
Distributional Quantum Query Complexity
Shalev Ben-David and M. H. Ebtehaj
Quantum query complexity enjoys a variety of pleasing joint computation properties: for example, a composition
theorem asserting $Q(f\circ g)=\Theta(Q(f)Q(g))$ for all Boolean functions $f$ and $g$; a direct sum
theorem asserting that computing $k$ copies of a function (or search problem) costs $\Omega(k)$ times as much as the cost of computing one copy; and a direct product
theorem asserting that for Boolean functions, even succeeding at the direct sum problem with exponentially small probability still requires $\Omega(k)$ times the cost of computing one copy to bounded error. However, all of these results are strictly for worst-case
quantum query complexity. For example, if we have a fixed distribution $\mu$ over inputs, the direct sum
theorem says nothing about the
quantum query complexity of computing $k$ copies of $f$ when the input comes from the product distribution $\mu^k$ instead of being worst-case. (Note that while a standard Yao-type minimax
theorem guarantees a hard distribution for the direct sum problem, there's no guarantee that this hard distribution is a product distribution.) A similar problem occurs for the direct product
theorem and the composition
theorem: none of these results respect distributions. In this work, we give distributional joint computation lower bounds for the composition, direct sum, and direct product problems. Along the way, we introduce some new tools for handling
quantum query lower bounds, including (a) a new “multiplicative” variant of the $\gamma_2$ norm (which we use in place of the multiplicative adversary method for proving the direct product
theorem), and (b) a new “Shaltiel-free” measure of
quantum query complexity, which we show characterizes the composition behavior of distributional
quantum query complexity and satisfies pleasing properties.
Read the full abstract ▾
Machine Learning · Cross-listed from stat.ML
Direct Intermediate Initialization for Tilted Diffusion Samplers
Gregory D. Bellchambers
Some
diffusion posterior samplers construct Gaussian-tilted intermediate distributions along the reverse process. We observe that these targets can be pulled back to clean-space posteriors with weaker conditioning, with samples transported analytically to the corresponding noisy-space target through a Gaussian bridge. For the sequential Monte Carlo (SMC) sampler MCGDiff, the effective observation variance of this pulled-back problem is up to twice the
diffusion-noise variance. We exploit this structure to initialize MCGDiff directly at an intermediate time: an approximate solver samples the softened clean-space posterior, the Gaussian bridge maps these samples to the tilted target, and only the remaining SMC suffix is run. This trades asymptotic consistency for finite-particle performance. With moment-matching posterior sampling (MMPS) as the solver, the hybrid improves sliced Wasserstein distance by roughly $2\times$ at matched particle count on a structured Gaussian-mixture inverse problem, and by more than an order of magnitude when the posterior-relevant mode is rare under the prior. A prior-initialization control, which retains the bridge but drops the clean-space conditioning, shows that on MCGDiff's standard Gaussian-mixture
benchmark most of the improvement is insensitive to the conditioning. Conditioning the initialization gives a further consistent gain on the structured problem, and becomes decisive on a rare-mode problem, where resampling cannot repopulate a mode absent from the initial population.
Read the full abstract ▾
Machine Learning
Towards Looped Models Done Right, Part II: Rethinking at Fixed Points
Benhao Huang, Chufan Shi, Junlin Chen et al. (+4 more)
Every recurrence of a looped
language model adds cost in training, decoding, prefill, and reinforcement learning (RL). The closer recurrent states get to fixed points, the less the path to them matters. This enables truncated backpropagation in training; terminal key-value (KV) sharing for decoding with almost no loss in accuracy; a distilled student that prefills up to 1.79x faster; and RL updates that compute gradients from saved rollout states, 2x faster than backpropagating through the replayed trajectory. We therefore improve the two components of training that shape these fixed points: the depth prior and input injection. Fixed-depth training breaks KV sharing, and Huginn's broad depth prior supports sharing but dilutes supervision at the target depth more than sharing requires; we learn the prior from prediction feedback, with an
entropy term that keeps it broad. Existing injection schemes let the state's component along the input amplify or cancel the injection; we remove this component with orthogonal injection. From 100M to 1.6B parameters, the learned prior and orthogonal injection lower perplexity at every scale relative to Huginn's prior and existing injection schemes, respectively. At 1.6B, the learned prior with a 3x smaller KV cache matches the downstream average of fixed-depth training with the full cache.
Read the full abstract ▾
All authors Benhao Huang, Chufan Shi, Junlin Chen, Shicheng Wen, Zhengzhong Liu, Eric Xing, Xuezhe Ma
Computer Vision
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
Computation and Language
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
Computation and Language
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
Computational Complexity · Cross-listed from quant-ph
Natural proofs for quantum state preparation lower bounds
Christine Li and Natalie Parham
We identify a barrier that helps explain why proving stronger
quantum state-preparation lower bounds has been so difficult. In particular, we establish a
quantum analogue of the Razborov-Rudich natural proofs barrier for state-preparation lower bounds. We call a property of
quantum states natural if it holds for a sufficiently large fraction of Haar-random states and can be efficiently tested when given all of the state's amplitudes. Under a standard cryptographic assumption, we show that no natural property can prove superpolynomial state-preparation lower bounds even against a fixed level of the Magic Hierarchy. We show that several existing state-preparation lower-bound techniques are natural in our sense, including arguments based on approximate degree, not being a unique ground state of a local
Hamiltonian, and mutual information.
Read the full abstract ▾
Computation and Language
PlotGround: Grounding Plot Digitization in Real Scientific Figures and Their Source Data
Yaohui Zhang, Binxu Li, Haoyi Duan et al. (+7 more)
Scientific figures often encode quantitative results that are not readily available in machine-readable form, making accurate plot digitization important for verifying and reusing published findings. Yet it remains unclear how accurately current models recover plotted values from real scientific figures, as existing
benchmarks rely largely on synthetic charts or cover only a limited range of chart types. We introduce PlotGround, an automated pipeline for building plot digitization
benchmarks from real scientific figures and their author-released source data. PlotGround maps figures to source tables, identifies reconstructable panels, and generates quantitative questions with source-grounded reference values. We use PlotGround to construct PlotGround-1k, a human-verified
benchmark of 1,119 questions from 1,066 bioRxiv preprints. Across sixteen
multimodal models, the best reaches 87.5% accuracy at a $\pm 5\%$ relative-error tolerance. Tightening the tolerance to $\pm 2\%$ lowers every model's accuracy by 11-24 percentage points, revealing a gap between approximate visual reading and precise quantitative recovery. PlotGround's paired figure-source structure lets us compare how accurately the same values are recovered from figures and from source tables. Providing source tables instead of figures raises a coding
agent's accuracy from 90.0% to 97.4% while cutting cost by 72%.
Read the full abstract ▾
All authors Yaohui Zhang, Binxu Li, Haoyi Duan, Jiacheng Miao, Yixin Wang, Xinran Du, Chenyue Li, Shilong Liu, Kevin Wu, James Zou
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 ▾
Machine Learning
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
Machine Learning
Private online learning and prediction for Littlestone classes
Amartya Sanyal
We study mistake bounds for differentially private online learning and online prediction under oblivious realisable adversaries. Online learning requires the learner to release a hypothesis at each time step whereas in online prediction, the learner only needs to make predictions without releasing a hypothesis. Using a novel lower bound for private online learning and an upper bound for private prediction, we show that the sample complexity of these two problems are separated by a factor that grows with the time horizon for every class of finite Littlestone dimension $d$. First, we prove that every $(\epsilon,\delta)$-private online learner has a deterministic realisable stream of length $T$ on which the mistake bound is at least $\mathbb E [M_T]= O\left(\frac d\epsilon \log^{2/3} T\right)$. In particular, this is the first non-trivial lower in the range $1/T<\delta<1/\log T)$ left open in earlier works[SR22,DSS24,LWY24]. Second, we prove that for every class of of Littlestone dimension $d$, there exists an $(\epsilon,\delta)$-jointly private predictor with at most $2^{2^{cd^2}}\epsilon^{-2}\log^2(2/(\epsilon\delta))$ expected mistakes, independently of $T$, for some absolute constant $c>0$. Thus, for every fixed class of finite Littlestone dimension when $\delta=\Theta(1/\log T)$, private learning requires $O\left(\log^{2/3} T\right)$ expected mistakes, whereas private prediction admits $O(\log\log T^2)$.
Read the full abstract ▾
Sound
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 ▾
Data Structures and Algorithms · Cross-listed from quant-ph
CV-QAOA: Efficient Low-Depth Quantum Optimization of Continuous Variables
Sriram Bharadwaj, Di Luo and Leo Zhou
We study a Continuous-Variable
Quantum Approximate Optimization
Algorithm (CV-QAOA) for high-dimensional continuous optimization. Our formulation extends an earlier CV-QAOA proposal with a variationally optimized initial state and recovers the convergence guarantees of
Quantum Hamiltonian Descent (QHD) in the high-depth limit. We prove rigorous performance guarantees of CV-QAOA on several families of cost functions. First, we show $d$-step CV-QAOA minimizes any $d$-dimensional strictly convex quadratic function with $2d$
quantum queries to the cost function. We then analyze a family of nonconvex "Rotated Double Well" (RDW) functions with $2^d$ local minima introduced by arXiv:2311.00811. While prior work showed QHD reaches its global minimum with $\tilde O(d^3)$ queries, we prove that 1-step CV-QAOA solves RDW with just two
quantum queries. Although general-purpose classical solvers need superpolynomial time for RDW and structure-awareness can reduce the cost to
polynomial time, we show that the 1-step CV-QAOA protocol can be efficiently dequantized, and that a gradient-aligned line search succeeds with $O(d)$ queries, nearly matching the information-theoretic $\Omega(d/\log d)$ query lower bound. To move beyond the dequantizable regime, we introduce a “Rotated Square Well” (RSW) problem, whose globally flat landscape suppresses useful local gradient information. For this family, we show that an adiabatic
evolution simulated by CV-QAOA can reach the global minimum using $d^{o(1)}$ queries. On the other hand, any classical
algorithm that learn the hidden rotation in RSW provably requires $\Omega(d^2/\log d)$ queries, a bound we nearly match with an explicit $\Theta(d^2\log d)$-query classical
algorithm.Numerical simulations on deflected corrugated spring and Easom functions illustrate the promising performance of CV-QAOA on more general problems.
Read the full abstract ▾
Computer Vision
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
Computer Vision
Less Context, Better Geometry: Masked Geometric Encoder for Robust 3D Foundation Models
Zhimin Shao, Xijun Liu, Zhaoliang Zhang et al. (+4 more)
Recent progress in 3D foundation models has enabled rapid 3D reconstruction and camera calibration by leveraging learned 3D priors from vast amount of spatial data. However, the all-to-all global
attention design leads to quadratic complexity and limits long-sequence inference; unconstrained cross-view interactions also can propagate unreliable evidence from occluded or visually similar but geometrically distant views. In this paper, We introduce a Masked
Geometric Encoder (MGE), which promotes the learning of robust
geometric representations under incomplete cross-view context. During training, MGE strategically drops frame
tokens from global
attention and distills from a pretrained full-context teacher model. This allows the model to learn an intrinsically richer per-frame representation while providing sufficient intermediate supervision to avoid performance degradation. Through extensive experiments, we show that MGE leads to much stronger performance under occlusion and doppelganger views while retaining high performance on standard
benchmarks. Such a richer frame representation also leads to more effective
token reduction during inference. To this end, we develop a novel Anchor-Guided Adaptive
token merging technique that preserves representative anchor frames while jointly merging redundant
tokens from the remaining views. Compared to other efficient inference approaches, we can achieve inference speedup while consistently maintaining higher reconstruction quality, particularly in limited-view settings.
Read the full abstract ▾
All authors Zhimin Shao, Xijun Liu, Zhaoliang Zhang, Yutao Tang, Abhay Yadav, Rama Chellappa, Cheng Peng
Data Structures and Algorithms · Cross-listed from quant-ph
Finding Gaussian Structure in Bosonic States
Alvan Arulandu, Sitan Chen, Ziyun Chen et al. (+2 more)
We study agnostic
tomography of pure bosonic Gaussian states: given copies of an arbitrary $n$-mode bosonic state $\rho$, the goal is to output a pure Gaussian state whose infidelity with $\rho$ is at most $\mathrm{opt} + \epsilon$, where $\mathrm{opt}$ is the minimum infidelity achievable by any pure Gaussian state. We give efficient protocols achieving this in both the high and low fidelity regimes. When $\mathrm{opt}$ is below some universal constant, our protocol has runtime and copy complexity which is strongly
polynomial in $n, 1/\epsilon$ and $\log \log E$, where $E$ is the energy of the closest pure Gaussian state. For arbitrary $\mathrm{opt}$, our protocol uses $(n+1)^{\mathrm{poly}(1/\epsilon)} \mathrm{poly}\left(1+\log\log(E)\right)$ copies and runtime. As a corollary, we obtain the first truly tolerant Gaussianity testing protocol for distinguishing whether $\mathrm{opt} > c + \epsilon$ or $\mathrm{opt} < c - \epsilon$, for any threshold $c\in(0,1)$. We also prove $\mathrm{poly}(n,1/\epsilon)$ runtime is impossible, unless $\mathrm{NP}\subseteq\mathrm{BQP}$. Our protocols follow a shared paradigm: first, we iteratively use general Gaussian measurements combined with techniques from classical robust statistics to obtain a good warm start estimate, then we leverage non-Gaussian measurements to refine this warm start using convex and non-convex optimization methods. Interestingly, we prove that non-Gaussian measurements are necessary to match the strong agnostic guarantees we obtain, and in fact these guarantees are provably superior to what is possible for robustly estimating classical Gaussians.
Read the full abstract ▾
All authors Alvan Arulandu, Sitan Chen, Ziyun Chen, Jerry Li, Eric Ma