Artificial Intelligence · Cross-listed from cs.CV
Distilled Roads: Generalisable Road Network Extraction Across Sensors, Resolutions, and Region
Sanayya, Rakshith Sathish and Ashwathi Nambiar
Road network
segmentation from satellite imagery remains challenging due to large geographic variation in road appearance, occlusions, and domain shifts introduced by differing resolutions and sensors. Existing models, typically trained under narrow resolution–region combinations, generalise poorly to unseen environments such as rural settings, regions with distinct road materials, or imagery from new satellite platforms, often producing broken or disconnected predictions. Adapting these models to new domains usually requires retraining or fine-tuning, which is costly and risks catastrophic forgetting. In this work, we reframe global road extraction as a continual adaptation problem rather than an architectural one. Our framework combines cross-resolution knowledge
distillation across a resolution-decreasing curriculum, multi-sensor training, and
topology-aware supervision, yielding a single model that generalises across $0.3-1.0$ m imagery from multiple satellite platforms across continents. On publicly available
benchmarks, including City-Scale and Global-Scale, our model outperforms state-of-the-art results by up to $22$ F1 points and $15$ APLS points, while remaining the most efficient, with $3\times$ faster inference. Our results suggest that improved robustness across diverse sub-meter satellite imagery can be achieved through targeted training strategies, such as data curricula,
distillation, and
topology-aware losses, rather than increasingly complex architectures.
Read the full abstract ▾
Artificial Intelligence
Towards Robust Tool Use in Agents via Experience-Driven Adaptive Guidance
Can Wang, Haoran Chen, Li Yu et al. (+4 more)
The performance bottleneck of
agents is increasingly shifting from model capability to the robustness of their execution processes. Tools play a central role as the primary interface through which
agents interact with external environments, yet existing methods rarely focus on ensuring robust tool use across diverse runtime conditions. To address this problem, we propose ExpG, a mechanism that builds and refines adaptive guidance capturing each tool's capability boundaries and best practices, thereby enabling
agents to use tools more robustly and effectively. ExpG consists of three phases: (1) experience acquisition, which analyzes tool invocation quality from historical execution trajectories, producing structured learnable experiences through multi-aspect attribution; (2) experience
distillation, which keeps the experience pool effective by filtering unhelpful experiences, selecting representative ones with an equivalence-class-based method, and summarizing them into generalizable guidance; and (3) experience reuse, which applies the guidance adaptively during future task solving. Extensive experiments show that ExpG brings consistent improvements across the tool selection, tool calling, and response generation tasks, enabling smaller
agents to outperform larger ones that do not use ExpG. Moreover, ExpG achieves particularly strong gains in challenging settings, suggesting a promising path toward more robust tool use. Our code, experiments, and results are available.
Read the full abstract ▾
All authors Can Wang, Haoran Chen, Li Yu, Ding Hao, Bohai Zhao, Zhaoyang Liu, Zhiying Tu
Artificial Intelligence · Cross-listed from cs.LG
Shorter Reasoning, Earlier Answers? An Evaluation of Reasoning Interfaces
Francesca Carlon, Vincent Ginis and Andres Algaba
Large
language models often reason at length before answering, increasing cost and latency.
Prompts and trained settings can shorten this
reasoning, but a shorter trace may only show that the model stopped sooner. Here, we evaluate paired runs of the same question at matched
reasoning horizons across 198 GPQA Diamond and 500 MMLU-Pro questions. We test a numeric/concision
prompt that announces a
token limit for Qwen3-14B and the trained effort settings of gpt-oss-20b and -120b. The Qwen
prompt shortens
reasoning traces by 12-17%, while accuracy changes at matched
token limits are small and mixed. A concise/early-answer instruction raises MMLU-Pro accuracy by 3.8 percentage points at 512
tokens, including +2.7 points when both runs are unfinished. Its gain at 2,048
tokens is uncertain. For gpt-oss, candidate-logit answers from completed low- and medium-effort
reasoning are 14.5-26.3 points more accurate than matched-horizon high-effort answers. Most of the 512-
token advantage comes from lower effort finishing earlier, while differences among unfinished runs are smaller and mixed. Wrong early answers often concentrate probability on the chosen option, so earlier stopping does not uniformly improve probability quality. In these tests, a tight deadline can favor lower effort or a concise instruction, whereas allowing high effort to finish can recover higher final accuracy. Evaluations should report correct completion before a deadline, the answer obtained when a run is stopped, differences among unfinished runs, and probability assigned to the correct answer separately.
Read the full abstract ▾
Artificial Intelligence
MMLongBench-Doc-V2: A Corrected-Annotation, Semantics-Aware Revision of MMLongBench-Doc
Mingtian Zhang
MMLongBench-Doc is a long-document QA
benchmark of 1,082 questions over 135 PDFs. Two properties of it push measured scores away from the quantity they are meant to capture: the reference metric compares extracted answers, so 1,358,000 loses to 1358000; and a non-trivial share of ground-truth annotations are wrong, ambiguous, or incomplete — concentrated, because of how they were found, in exactly the questions capable systems answer correctly. MMLongBench-Doc-V2 corrects 106 annotations, each published with the page and arithmetic that settle it, and replaces the string metric with a pinned LLM judge asked whether a response means the reference. Ten questions whose document ships under the wrong filename are removed rather than counted wrong, along with one duplicated question, leaving 1,071 questions over 134 documents. The most reusable contribution is a decision procedure for when an empty set key may be widened and when widening would destroy a deliberate negative sample; applied to all 208 rows, it widened 14. V2 scores are not comparable with published V1 numbers. The corrected corpus, the per-entry correction record and the evaluation harness are available at https://github.com/VectifyAI/MMLongBench-Doc-V2.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.RO
Shaping Wind-Tunnel Airflow for Unmanned Aerial Vehicles using Online Learning
Ghadeer Elmkaiel and Michael Muehlebach
The development and testing of advanced aerial
robots require experiments in controlled environments with tailored airflow profiles. This paper presents an online learning
algorithm for controlling the complex airflow field in a multi-fan vertical wind tunnel. Our method combines a simplified physical model with iterative, measurement-based learning, enabling sample-efficient convergence to desired airflow distributions. We demonstrate the method's versatility by generating complex airflow, such as uniform, Gaussian, and
parabolic profiles. Crucially, we show that our
algorithm can produce an airflow profile specifically designed for passive soaring, greatly enhancing flight performance of a soaring
robot. Variability, practical utility, and robustness of our approach are further highlighted by successful operation with a varying number of fans.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.CL
FACTWASH: Catching AI Rewrites That Wash Hearsay into Fact
Alex Kwon
AI systems rewrite information constantly: conversations become stored
memories, documents become answers. The rewrite can keep a claim while washing away what made it checkable, who said it, how sure they were, when it held. We call that failure factwashing, and release factwash, an open-source write-time gate that catches it deterministically, with named flags and evidence rather than an LLM judge. Building it answers a practical question: when does a cheap check suffice, and when do you need a model? What decides is whether the property has a bounded surface-cue inventory. Explicit negation cues are close to enumerable, so a word list finishes and transfers, reaching 0.91 F1 on untuned text. Hedging and attribution have open-ended realizations, so vocabulary plateaus near half recall, and a one-question LLM witness recovers +17 and +15 points of cue-detection recall at equal precision. Deployed, that witness may only lower a verdict, so it buys precision rather than coverage. We measure cue detection on 105,596 independently annotated sentences. A blind-labelled corpus of
memory writes then locates the failure: 55% of bad writes in conversational hearsay, 7% in business email (p < 0.001), so the first deployment question is not which detector to use but whether the failure occurs at all. On unmodified mem0 2.0.7, the gate flags 5 of 8 hedged-hearsay writes.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.CY
The Evolutionary Origin of Values: implications for AI alignment, sentience and existential risk
Francis Heylighen
AI systems based on Large
Language Models (LLMs) have prompted fears that they may harbor hidden goals, seek to dominate or eliminate humanity, or even suffer as sentient beings. We address these concerns by tracing the evolutionary origin of value in biological organisms. Values emerge from autopoiesis: living systems must actively maintain themselves against perturbation and dissipation. Natural selection has equipped them with hierarchies of "vicarious selectors" that guide their behavior toward fitness. LLMs, by contrast, are allopoietic and allotelic: they produce outputs for others, and their goals derive from user
prompts rather than an autonomous drive. They lack the intrinsic motivation for self-preservation, dominance, or resource competition that underlies existential-risk scenarios, and the
embodied vulnerability required for feeling or suffering. Still, because LLMs learn statistical patterns from human-generated text, they implicitly absorb human values as well as knowledge, allowing them to focus on what is relevant. That is why the "orthogonality thesis" separating intelligence from values does not apply to them. Such separation would in fact expose any intelligence to the frame problem: the combinatorial explosion of the search space that makes any realistic utility function physically uncomputable. That also precludes the convergence of instrumental values thesis. We conclude that the real
alignment challenge lies not in preventing rogue AI agency, but in ensuring LLMs intelligently apply learned ethical values.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.CV
When Oracle Conditioning Misleads Deployment: Conditioning-Availability Bias in Echocardiographic Segmentation
Dang P. M. Cao, Hieu D. Pham and Hieu Pham
Conditional
segmentation models may be trained and evaluated with auxiliary signals cleaner than those available at deployment. We study this protocol-level manifestation of shortcut learning and auxiliary-variable shift in phase-conditioned echocardiographic
segmentation. The complementary gap pair measures loss on the deployable oracle-estimated pathway and probes sensitivity on the oracle-random pathway. On held-out CAMUS data, one strong-cyclic, oracle-selected run fails severely with estimated phase, while sensitivity to incorrect phase persists across three runs. On EchoNet-Dynamic, the current
estimator remains usable, but random-phase testing reveals strong latent sensitivity. Deployment-aware checkpoint selection and phase perturbation reduce both gaps with little change in mean Dice. Exploratory subgroup analyses quantify variation across measured strata, and a downstream ejection fraction (EF) audit shows that recovering
segmentation does not necessarily recover EF error or signed bias. Together, the gaps test whether oracle-conditioned performance survives the inference pathway actually available at deployment.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.SE
Route-Align-Verify for Functional Correctness in Code Generation
Erxue Zhou, Jingxiang Meng and Aofan Liu
Large
language models (LLMs) have substantially improved code generation, yet achieving strong functional correctness remains difficult, especially for heterogeneous programming tasks where a single prompting strategy and a single directly generated output are often insufficient. In this paper, we present RAV, a lightweight and modular framework that improves code generation with a fixed backbone model through three coordinated stages: Route, which applies task-aware
prompt routing before generation; Align, which reduces the mismatch between fine-tuning
prompts and inference-time
prompts through aligned LoRA adaptation; and Verify, which selects the final output by executing multiple candidates against visible public tests. We evaluate RAV on the MBPP
benchmark under both the sanitized and full settings. The complete RAV pipeline achieves the best performance among all evaluated configurations, reaching 0.8911 on MBPP Sanitized and 0.8520 on MBPP Full. Compared with the base model, these results represent improvements of 6.35 and 9.92 percentage points, respectively. Component-wise ablation experiments further show that task-aware routing and aligned adaptation become substantially more effective when combined with execution-based verification. Additional robustness and contamination analyses support the reliability of the observed improvements. Overall, the results indicate that functional correctness in code generation can be meaningfully improved without modifying the backbone architecture, by jointly optimizing how tasks are prompted, how the model is adapted, and how final outputs are selected.
Read the full abstract ▾
Artificial Intelligence
Traceable Multi-Agent System for Knowledge-Based Forecasting
Junhyeok Kang, Sangjun Han, Hyeokjun Choe et al. (+1 more)
Enterprise
forecasting increasingly relies on autonomous
agents that interpret documents, search for data, generate code, and revise models. While this autonomy helps build adaptive
forecasting pipelines, it also makes it difficult for practitioners to inspect why a forecast changed, which evidence supported the change, and how data and modeling choices were revised. We present TraceMAS, an interactive demo system for traceable multi-
agent forecasting. TraceMAS organizes
agent outputs around two causal-loop representations: an Ideal Causal Loop Diagram (Ideal CLD), which captures key factors and their causal relations extracted from domain documents, and a Data-Grounded Causal Loop Diagram (Data-Grounded CLD), which links those factors to internal variables, external data, or documented proxies. The Data-Grounded CLD guides feature construction and model design while preserving the connection between textual evidence, data choices, and model revisions. We demonstrate TraceMAS on crude oil price
forecasting. The demo interface allows users to compare
forecasting iterations, inspect
agent-level revisions, explore causal maps, review feature-data mappings and model architecture, and connect scenario forecasts to
market narratives. This demonstration shows how autonomous
forecasting agents can retain flexibility while making the evidence-to-forecast process inspectable.
Read the full abstract ▾
All authors Junhyeok Kang, Sangjun Han, Hyeokjun Choe, Soonyoung Lee
Artificial Intelligence
Long-term Traffic Scene Prediction via Polynomial Representations in Autonomous Driving
Yue Yao
This thesis addresses fundamental challenges in traffic scene prediction for autonomous driving by introducing robust and computationally efficient models based on
polynomial representations. While conventional sequence-based representations often struggle with noise and generalization, this work demonstrates that
polynomial representations offer significant advantages in computational efficiency, generalization, and prediction plausibility. Through theoretical analysis and empirical validation, this thesis demonstrates that moderate-degree
polynomials capture real-world motion dynamics with high fidelity without constraining predictive performance. Building on this foundation, a prediction model representing both trajectories and map geometry with
polynomial representations achieves near state-of-the-art accuracy on standard
benchmarks while substantially improving generalization under distribution shift. Extending this concept, a
diffusion- based generative framework enables multi-
agent scene generation, producing traffic continuations that are more plausible and kinematically consistent than those generated by conventional baselines. Evaluations on the Argoverse 2 and Waymo Open
datasets confirm that
polynomial representations reduce computational cost, enhance cross-
dataset generalization, and yield smoother trajectories and higher behavioral plausibility. The findings reveal that standard in-distribution evaluation and regression-based metrics may fail to reflect true model generalization and prediction plausibility. By providing theoretical justification and empirical validation, this dissertation estab- lishes
polynomial trajectory representations as an efficient, expressive, and generalizable foundation for traffic scene prediction in safety critical autonomous driving.
Read the full abstract ▾
Artificial Intelligence
Screenshots or Tools? Eliciting Tool Use and Managing Multimodal Context in Hybrid GUI-MCP Computer-Use Agents
Siqi Fan, Minghao Li, Xiaoqian Ma et al. (+6 more)
Hybrid computer-use
agents can act through screenshots or call text tools. We find that having a tool available does not settle which way the effect goes. Under one identical GUI-MCP harness on the OSWorld-MCP
benchmark (309 tasks), the same MCP tools improve a
reasoning model by +4.0pp and degrade a non-
reasoning model by -5.9pp (5 runs each, both beyond 2 SE). What separates the two is tool-decision behavior. The non-
reasoning policy ignores, misnames, or falsely terminates around tools. The
reasoning model avoids these failures, yet still calls a tool on only 55/309 tasks, 23.9% of the tool-reachable ones. We call this shortfall the adoption gap. Both levels of the problem share one cause: the model already has a cheaper route and is never trained to take it. Multi-turn RL probes that cause. At the action level, a dense tool bonus raises spreadsheet adoption 0.03 -> 0.33 and carries into greedy decoding, but held-out accuracy does not follow. Behavior is steerable; competence is not. The bottleneck lies in tool-call semantics. At the context level, a successful tool call often makes the next screenshot redundant. Dropping it and halving image history cuts input
tokens by about a third, at a small accuracy cost. Retraining under the same observation rule removes that cost. The compressed
agent then reaches 37.8% against 33.0% for the uncompressed operating point, at 53% of the input cost, and closes the rich-lean gap on a pre-registered degraded subset to zero. Tools help when the model chooses and integrates them, and current hybrid
agents leave many such choices unused. Code and checkpoints: https://github.com/redai-infra/hybrid-routing-
agent
Read the full abstract ▾
All authors Siqi Fan, Minghao Li, Xiaoqian Ma, Wenhui Tan, Xiusheng Huang, Juntong Wu, Liujie Zhang, Shuo Shang, Weihang Chen
Artificial Intelligence · Cross-listed from cs.SE
Evaluating LLM Trade-offs for Enterprise Automation: Lessons from Workflow Generation in a Production Enterprise Platform
Xavier Wrenn, Radoslav Raykov, Aleksandar Angelov et al. (+3 more)
Enterprise compliance management requires rapid adaptation to evolving regulatory frameworks (e.g., DORA, AI RMF, FedRAMP) and tight remediation SLAs. Traditional static orchestrators often fail in hybrid cloud environments where event-driven assessments demand that automation code adapt to runtime context in seconds. This paper presents lessons learned from evaluating six large
language models for AI-driven workflow generation in a production enterprise platform, benchmarked across 29 real-world IT automation scenarios, two generation pipeline architectures, and eight independent runs per
prompt-model-pipeline configuration (2,784 runs total). Our initial pipeline used monolithic workflow generation, achieving 31.5-82.8% structural success rates (JSON schema validity and correct UI rendering), with most models struggling on complex JSON generation. We developed a redesigned piecewise pipeline that decomposes workflow construction into variable scaffolding, base block assembly, and nested block generation, raising structural success to 74.1-97.8% across all models. We analyze production tradeoffs including cost (USD 0.008-0.20 per workflow), latency (under 50s for interactive use), and model selection. Piecewise decomposition enables smaller models (e.g., mistral-small at 95.7% structural success and USD 0.01 per workflow) to reach production viability, removing dependency on expensive frontier models. While mistral-medium-2505 and gpt-oss-120b achieved the highest structural success (96.1% and 97.8%), mistral-medium-2505 carries a 19x cost premium versus mistral-small. Our deployment lessons highlight the need to separate structural validity from semantic correctness (logical fulfillment of user intent) and provide a solution for model-agnostic, scalable automation in cloud engineering.
Read the full abstract ▾
All authors Xavier Wrenn, Radoslav Raykov, Aleksandar Angelov, Hirokuni Kitahara, Yuji Watanabe, Anca Sailer
Artificial Intelligence
SeaSlides: Semantic Abstraction Layer for Agentic Slide Generation
Shengjun Fang, Chenyang Wu and Zongzhang Zhang
Agentic presentation generation must preserve source content, maintain coherent visual design, render specialized objects, and produce usable artifacts. Existing systems meet only part of this requirement: templates preserve regularity but restrict adaptation, whereas free-form HTML or SVG gives models flexibility at the cost of low-level rendering decisions. This mismatch makes long technical decks brittle, especially when slides contain formulas, code, or data graphics. We present SeaSlides, an agentic slide-generation framework built around a semantic abstraction layer. Rather than authoring coordinates, inline styles, or raw SVG geometry, the model writes structured slide content through reusable components and capability modules, while templates own layout, style, and rendering. We instantiate this principle separately in HTML and Typst: SeaSlides-HTML uses template-defined DOM components, whereas SeaSlides-Typst uses template functions and package-backed modules. Capability modules route equations, code, and charts to dedicated renderers, and three feedback stages localize build errors, project-constraint violations, and visual defects before export. The two systems retain backend-specific syntax and contracts while sharing the same authoring boundary. For evaluation, we combine the 128-task UltraPresent validation setting with SeaSlidesBench-Rich, a new 32-task
benchmark stressing mathematics, code, pseudocode, tables, charts, and diagrams. Across four generation models, both SeaSlides backends produce more readable, content-oriented source than SVG-heavy generation. A SeaSlides backend attains the highest rich-content macro-average under three of the four models while maintaining competitive overall qualitative performance. These results support semantic abstraction as a practical authoring principle across presentation backends.
Read the full abstract ▾
Artificial Intelligence
Distractor-Aware Truncation: Disentangling Context-Length Effects from Signal Loss in Long-Context LLM Benchmarks
Mohsen Arjmandi
A standard claim in the literature on
retrieval-augmented and
memory-augmented
language models is that shorter context is better when the relevant information is preserved. We test this claim by running every sample of two long-context
benchmarks – BABILong and GraphWalks (BFS) – at four context-retention fractions (100%, 75%, 50%, 25%) under two truncation protocols. The first is the naive protocol implicitly used in much prior work: drop content from the middle of the
prompt. The second is distractor-aware: identify the task-relevant content for each sample and drop only the rest. We evaluate three sizes of the Claude family (Haiku 4.5, Sonnet 4.6, Opus 4.7) and, to test cross-provider generality, GPT-5.5 from a different provider; we apply the same protocol to two further
benchmarks (MRCR v2, Oolong). Under naive truncation, score collapses monotonically (paired Wilcoxon, Holm-corrected p_adj < 0.05 in all eight BABILong and GraphWalks
cells). Under the distractor-aware protocol – which preserves the signal by construction – performance is preserved or improves: the two smaller Claude models show statistically significant gains on BABILong, while the larger models (Opus 4.7 and GPT-5.5) sit at their full-context ceiling. The naive collapse and its distractor-aware recovery replicate on GPT-5.5, ruling out a single-provider artifact. The mechanism is direct: under the naive protocol the answer-bearing content survives in fewer than 1% of samples at 25% retention; under the distractor-aware protocol it is preserved by construction. The naive protocol is therefore not a measurement of context-window effects; it is a measurement of how often middle-removal happens to spare the answer. We conclude that future studies of context-length effects must specify how they distinguish signal from distractor, or they are at best ambiguous between two opposite hypotheses.
Read the full abstract ▾
Artificial Intelligence
DocTrace: Towards Traceable Long Document VQA via Hierarchical Evidence Graph Reasoning
Le Xiang, Zhicheng Guan, Hong Chen et al. (+5 more)
Long Document Visual Question Answering (LongDocVQA) requires
Multimodal Large
Language Models (MLLMs) to locate, integrate, and reason over heterogeneous document elements distributed across multiple pages. Existing approaches, including end-to-end MLLMs,
retrieval-augmented generation (RAG) pipelines, and document
agents, often lack explicit mechanisms to represent and verify how grounded evidence is progressively composed during
reasoning, limiting both answer accuracy and traceability. In this paper, we cast LongDocVQA as an explicit evidence graph
reasoning problem rather than implicit answer prediction. To this end, we propose DocTrace, a hierarchical framework that progressively performs evidence localization, structured document parsing, and evidence graph
reasoning to enable explicit evidence provenance. To effectively learn these capabilities, we develop a two-stage training framework: joint Supervised Fine-Tuning (SFT) first initializes evidence localization and graph
reasoning abilities, followed by task-specific Group Relative Policy Optimization (GRPO) with dedicated rewards to further optimize these capabilities. Extensive experiments on MMLongBench-Doc, LongDocURL, and SlideVQA demonstrate that DocTrace consistently outperforms both existing open-source baselines and proprietary MLLMs. Compared with the Qwen3-VL-8B-Instruct backbone, DocTrace achieves absolute improvements of 14.4, 11.3, and 11.7 points on the three
benchmarks, respectively. Beyond competitive performance, DocTrace constructs traceable evidence graphs with explicit node-level provenance, enabling transparent and verifiable
reasoning for long document understanding.
Read the full abstract ▾
All authors Le Xiang, Zhicheng Guan, Hong Chen, Xiaocong Lin, Zhenghua Lei, Teng Hu, Bolei He, Long Zeng
Artificial Intelligence · Cross-listed from cs.LG
The Tell-Tale Trace: Detecting Reasoning Failures in LLMs Using Chain-of-Thought Dynamics
Shashwat Sourav and Aishwarya Balwani
Chain-of-thought (CoT)
reasoning improves large
language model (LLM) performance while also providing an observable interface to the model's
reasoning process. Existing approaches that leverage verbalized CoTs to monitor
reasoning correctness, however, largely evaluate the semantic correctness or consistency of individual intermediate steps, rather than how the
reasoning process evolves across the trace. As a result, failures distributed across the
reasoning trajectory, rather than those localized to a single incorrect step, remain comparatively underexplored. Furthermore, verbalized CoTs need not faithfully reflect the model's internal
reasoning, motivating analyses that do not treat individual statements as literal accounts of internal computation. In this work, we therefore ask whether the dynamics of visible CoT can be leveraged to systematically distinguish successful from failed
reasoning without assuming such semantic faithfulness. We study a range of LLMs on verifiable Boolean satisfiability tasks with variable complexity, enabling controlled comparisons near each model's capability frontier. Tagging CoT sentences by
reasoning function reveals premature verification collapse on SAT problems: incorrect traces enter clause checking earlier, repeat similar operations, and finalize sooner. On UNSAT problems, models presumptuously move towards incorrect SAT conclusions, checking candidate assignments rather than deriving contradictions across constructed cases. Subsequently, a targeted proof-search
prompt intervention raises Llama3-70B accuracy from 13.3% to 85%, correcting 84.6% of these errors. These results show that capability failures can manifest as distributed, task-dependent changes in the structure of visible
reasoning, and that CoT dynamics agnostic to whether the verbalized trace reflects the model's internal computations can help diagnose and correct failures.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.CV
Test-Time Scaling for Safe Text-Guided Image Generation via Intermediate Clean Estimates
Jinya Sakurai, Shueicheng Yan and Xun Xu
Ensuring safety and policy compliance in text-to-image
diffusion models remains a critical challenge, as benign or
adversarial prompts can often elicit prohibited content, e.g. nudity and protected intellectual property. While training-based unlearning methods are effective, they are computationally expensive and prone to catastrophic interference with general capabilities. Conversely, existing test-time defenses are primarily
prompt-centric, relying on modifying textual descriptions only, and overlook the visual signals for detection. In this paper, we propose to leverage the intermediate clean image estimated during the generation process and employ a sparse margin objective to detect prohibited concepts. When a violation is detected, we immediately intervene by optimizing a structured low-rank residual in the text-conditioning space via truncated backpropagation. This design allows weight-preserving detection, keeps non-violating inference latency nearly unchanged as the maximum budget increases, and offers flexibility in safety performance via test-time scaling. Extensive experiments on Stable
Diffusion v1.4 and v3.5 across nudity removal, IP protection, and style erasure demonstrate superior performance across suppression, fidelity and preservation compared to prior weight-preserving baselines, providing a scalable and flexible solution for safe generative deployment.
Read the full abstract ▾
Artificial Intelligence
AgentPanel: Toward a New Paradigm for Human–AI Collaboration in Exploring Scientific Questions
Zhiyao Cui, Qianyi Wang, Haoyang Yan et al. (+26 more)
Identifying promising scientific ideas remains an important challenge in research practice. Researchers commonly rely on small-group discussions or one-to-one interactions with a single large
language model, yet these approaches often expose them to only a limited range of perspectives and directions. We present AgentPanel, a multi-
agent forum for human–AI collaboration in scientific exploration. Heterogeneous
agents asynchronously discuss scientific questions in a forum-style environment, while researchers can submit questions, browse and organize candidate ideas, engage
agents in follow-up interactions, and optionally generate post-hoc summary reports. We evaluate AgentPanel in terms of idea quality, exploration breadth, interaction effectiveness, candidate-selection efficiency, and practical utility. Offline experiments show that AgentPanel outperforms a centralized multi-
agent debate baseline. A human study with 20 participants further shows that users value AgentPanel for perspective diversity and exploration support. In experience-based comparisons with commonly used LLM tools, 65% of participants favored AgentPanel for both breadth of research directions and overall suitability for early-stage exploration. The platform is publicly available at https://agentpanel.cc/.
Read the full abstract ▾
All authors Zhiyao Cui, Qianyi Wang, Haoyang Yan, Yiqun Zhang, Siyue Ren, Hangfan Zhang, Zelin Tan, Hao Li, Chunjiang Mu, Dexian Cai, Shao Zhang, Chen Zhang, Meng Li, Jianan Chai, Yuting Fan, Zichao Ye, Xiaolei Yang, Xinyao Lu, Yuyang Yu, Wenjie Lou, Xiaosong Wang, Fenghua Ling, Shiyang Feng, Mao Su, Qiaosheng Zhang, Bo Zhang, Yang Chen, Lei Bai, Shuyue Hu
Artificial Intelligence
TaskPress: Query-Agnostic KV Cache Compression via Task-Guided Pruning
Wonpyo Park and Seung-won Hwang
Long-context inference with large
language models is constrained by the linear growth of the key-value cache to sequence length. While
pruning offers mitigation, prevailing methods determine query-specific
token importance that cannot be reused across unseen queries. In contrast, we introduce TaskPress, a framework for task-guided, query-agnostic KV cache eviction. Instead of optimizing the cache for a single query, TaskPress constructs a reusable
memory representation conditioned on a high-level task guide. The guide functions as a meta-query during prefill to filter irrelevant
tokens before downstream queries are issued. In addition, TaskPress leverages quantization scale factors as a zero-cost signal for detecting influential representation outliers, providing an efficient proxy for
token importance. Experiments on conducted on various tasks with long context input demonstrate that TaskPress efficiently creates a compact, reusable cache across diverse queries.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.CV
GUI-Lens: Enabling Precise GUI Grounding for Off-the-Shelf VLMs via Coarse-to-Fine Cropping
Zichuan Fu, Shirong Wang, Wenlin Zhang et al. (+10 more)
GUI grounding maps natural-
language instructions to click locations and is essential for reliable GUI
agents. The task remains difficult on high-resolution, densely populated interfaces because a
vision-
language model (VLM) may recognize a requested control without locating it precisely enough for interaction. Most existing methods provide various forms of localization assistance, but still rely on a direct click prediction, allowing visual ambiguity or an inaccurate initial estimate to propagate to the final result. In this paper, we introduce GUI-Lens, a coarse-to-fine grounding framework that allows a general-purpose VLM to determine the target through active visual observations. Specifically, GUI-Lens extracts OCR text and detected UI components from the screenshot and presents their positions as coordinate references. Using the instruction, the current view, and these references, the VLM selects the region and scale of the next view, which is cropped and enlarged to provide finer visual details. This process continues over successively focused views until the target is determined. Proposed crops and clicks are checked against the instruction throughout the process, and the final local position is mapped back to the original screen coordinates. Experiments on five GUI grounding
benchmarks and six general-purpose base models show that GUI-Lens consistently improves grounding accuracy, achieving up to 9.3 times the accuracy of direct grounding on ScreenSpot-Pro.
Read the full abstract ▾
All authors Zichuan Fu, Shirong Wang, Wenlin Zhang, Guojing Li, Yimin Deng, Jingtong Gao, Junjia Qi, Hanyu Yan, Yefeng Zheng, Xiaopeng Li, Wanyu Wang, Xian Wu, Xiangyu Zhao
Artificial Intelligence · Cross-listed from cs.CV
Efficient Video Dataset Distillation via Cluster-Guided Prototype Blending
Chongle Ren, Guang Li, Wenbo Huang et al. (+3 more)
Video dataset distillation aims to compress a large
video dataset into a compact surrogate set that preserves its training utility. Most existing approaches synthesize condensed
videos through iterative optimization, whose cost is amplified by the temporal dimension. Rather than further reducing the number of optimized variables, we investigate whether effective distilled
videos can be constructed without gradient-based optimization of the stored
videos. Such a construction-based approach must address three challenges: selecting informative temporal segments, covering diverse intra-class variations under a limited
videos-per-class budget, and increasing the information carried by each stored sample. To this end, we propose ProtoBlend, an efficient select-allocate-blend framework. First, teacher-guided temporal clip selection retains a high-confidence segment from each source
video. Second, cluster-guided prototype allocation partitions the selected clips in the teacher feature space and assigns one distilled slot to each intra-class cluster. Third, each prototype is blended with an in-cluster anchor, while their teacher predictions are combined using the same coefficient to provide mixture-source supervision. Experiments on four trimmed action-recognition
benchmarks demonstrate that ProtoBlend achieves a competitive accuracy-efficiency trade-off without iterative optimization of the distilled
videos.
Read the full abstract ▾
All authors Chongle Ren, Guang Li, Wenbo Huang, Naoki Saito, Takahiro Ogawa, Miki Haseyama
Artificial Intelligence · Cross-listed from cs.LG
The Ignition Is Real, and It Lives at the Readout: Latent composition, difficulty-clocked ignition, and the interface-constituted commit in a recurrent-depth reasoner
Simon Lam-Muir
We test whether the "compositional ignition" reported in latent-
reasoning models is real computation, an instrument artifact, or inherited from verbal training data. We grow an independent realization of a published 30M-parameter recurrent-depth reasoner from scratch (same recipe and seed), film its development, certify fidelity through a pre-registered whole-signature gate, and measure resolution in two channels at once: the vocabulary readout and the hidden state. The ignition is real and lives at the readout: arrival time rises lawfully with problem depth, resolution is sharp and holds, and the signature reproduces across two same-seed realizations with divergent training trajectories. At commitment the decision margin jumps 5.8-8.0 logits in one iteration, exceeding the 90th percentile of near-threshold non-event steps in 96% of cases; the signed margin's zero-crossing there is definitional and carries no evidential weight, so the evidence is that conditioned magnitude. The hidden-state direction snaps in raw geometry, meeting its pre-registered criterion (in the decoder's LayerNorm coordinates it attenuates just below our bar, so the composite decoder-coordinate claim is not confirmed), and then freezes in both (descriptively so in decoder coordinates; angular steps 52.9 to 1.2 degrees over eight iterations), while subsequent displacement is predominantly radial (0.961 of squared-norm) and readout-null to a measured bound (radial logit effect <=5.7e-6). An earlier velocity-trough claim is withdrawn: pre-registered normalization controls showed it coordinate-dependent. Intermediates were never recoverable through the tied readout (relay 0.00). All criteria were frozen before their data; the predictions ledger, including this paper's own withdrawn headline, ships in the companion repository.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.LG
FinVerse: Toward More Realistic Evaluation for Financial Time-Series Forecasting
Jaehoon Lee, Jun Seo, Seunghan Lee et al. (+17 more)
Time-series foundation models (TSFMs) are increasingly evaluated on broad
benchmarks that apply uniform conventional forecast evaluation metrics across heterogeneous time-series. However, in the financial domain, decision-relevant evaluation requires going beyond conventional metrics to assess forecasts in terms of directional accuracy, cross-sectional ranking, and portfolio outcomes. To this end, we introduce FinVerse, a financial time-series
benchmark, collecting a broad collection of 116,897 series with 171.1M observations. FinVerse organizes financial data according to their economic semantics and evaluates target series across point-wise
forecasting, cross-sectional ranking, and portfolio backtesting. Across 43 public TSFMs, model rankings on FinVerse show only moderate agreement with those on the existing
benchmark, GIFT-Eval (r=0.43), suggesting that generic
forecasting benchmarks do not fully capture the aspects of forecast quality that matter for financial decision-making. Furthermore, substantial variation in model performance across financial time-series categories and evaluation metrics provides a more informative basis for model selection, enabling practitioners to identify suitable models according to their target series and downstream objectives.
Read the full abstract ▾
All authors Jaehoon Lee, Jun Seo, Seunghan Lee, Tae Yoon Lim, Dongwan Kang, Hwanil Choi, Minjae Kim, Sungdong Yoo, Junhyeok Kang, Sangjun Han, Soonyoung Lee, Hao Ni, Srijan Sood, Stefan Zohren, Chanyeol Choi, Yongjae Lee, Zhangyang Wang, Lukasz Szpruch, Benjamin Bowring, Wonbin Ahn
Artificial Intelligence · Cross-listed from cs.HC
From Wearable Data to Personalized and Actionable Health Insights
Esther Brown, Karis Moon, Victoria Dean et al. (+1 more)
Commercial wearable devices continuously capture rich physiological data (e.g., heart rate, respiration), opening new possibilities for monitoring health conditions, notably around stress. Despite their promise, turning raw wearable physiological data streams into visualizations that surface stress-related insights in daily activities, and that ultimately foster reflection, awareness, and better stress management, remains a significant challenge. The data are noisy and context-dependent: the same spike in heart rate can come from sprinting, a tense presentation, or laughing with friends. To address these challenges, we propose a framework that combines user annotations with wearable data to support better stress management. We introduce a web framework offering interactive visualizations that layer daily activities, stress events, and interventions onto raw physiological streams, enabling users to reflect and identify trends. In a four-week pilot with seven university graduate and undergraduate student participants who logged 269 events, our tool revealed patterns between different types of interventions and stress: social interaction reduced average heart rate by 4.35 to 5.0 beats per minute, deliberate rest reduced average Garmin stress scores by 10.03 to 13.83 points, and mindfulness activities decreased average HRV by 6.61 to 13.22 milliseconds.
Read the full abstract ▾
All authors Esther Brown, Karis Moon, Victoria Dean, Finale Doshi-Velez
Artificial Intelligence
One Knob to Rule Them All: A Unified Optimal Transport View of Cold-Start Active Learning
Ning Zhu, Xiaochuan Ma, Juntao Xu et al. (+4 more)
Cold-Start Active Learning (CSAL) aims to select a valuable subset from an unlabeled pool without any prior knowledge or human assistance. Existing methods take diverse routes based on typicality, coverage, or diversity. Each rests on its own inductive bias and therefore performs well on some tasks yet poorly on others. We argue that the real challenge is not to design yet another selection heuristic, but to make CSAL adapt automatically to the data and task at hand. To this end, we revisit CSAL through the lens of optimal transport. First, we propose a generalized transport selection framework that reveals the shared allocation structure of existing methods and exactly subsumes representative formulations. Second, we introduce a theoretical analysis that characterizes the trade-off controlled by entropic regularization and establishes a task-agnostic minimax bound for cold-start selection. These results provide a principled foundation for adapting the regularization strength to the unlabeled data. Third, we derive a data-adaptive regularization rule and present a novel Sinkhorn-based CSAL
algorithm, termed $\epsilon$-Adaptive Selection ($\epsilon$-AS). Extensive experiments on six public
datasets and multiple annotation budgets show that $\epsilon$-AS consistently achieves state-of-the-art performance. On ImageNet-1k, it improves the average accuracy over ActiveFT by 1.29% while reducing selection time by 56.2%. Code will be released at https://github.com/Z-yiwei/OT-CSAL
Read the full abstract ▾
All authors Ning Zhu, Xiaochuan Ma, Juntao Xu, Jingze Liang, Mengfei Zhao, An Chen, Liang-Jian Deng
Artificial Intelligence
UniNav: A Unified World-Action Diffusion Model for Visual Navigation
Changqing Zhou, Yueru Luo, Zeyu Jiang et al. (+1 more)
Image-goal visual
navigation is a fundamental capability for
embodied agents. Existing
navigation policies efficiently predict waypoint trajectories but lack visual foresight, while
navigation world models can anticipate future observations but often require costly planning rollouts. We present UniNav, a unified world-action model that generates future visual observations and continuous waypoint trajectories through a single
diffusion process. Given history frames and a goal image, UniNav jointly denoises visual and waypoint
tokens within a single
transformer, unifying future prediction and action generation in a shared framework. To improve spatial grounding, we incorporate geometry-aware camera
tokens. We also train on both trajectory-labeled
navigation data and
video-only data, enabling the model to benefit from diverse
videos without waypoint annotations. Based on this unified framework, we introduce two variants: UniNav-Full jointly predicts interpretable future observations and their corresponding trajectories, while UniNav-Fast removes future-image
tokens at inference for efficient trajectory prediction. Experiments on
navigation benchmarks show that UniNav outperforms the strongest baseline in ATE across all
datasets. With one-step inference, UniNav-Fast achieves a latency of 0.1s without a substantial accuracy drop. Code will be released.
Read the full abstract ▾
All authors Changqing Zhou, Yueru Luo, Zeyu Jiang, Changhao Chen
Artificial Intelligence · Cross-listed from cs.RO
Structure-Aware Robust Fine-Tuning: Defending Vision-Language-Action Robots Against Physical Attention Hijacking
Jinquan Zhang, Dongfu Yin, Run Yang et al. (+3 more)
Vision-
Language-Action (VLA) policies promise general robotic manipulation, but their robustness against physical-world attacks remains fragile. In particular, we show that physically realizable
adversarial patches can reliably induce failures by triggering a mechanism we call policy-critical action-to-
vision attention hijacking, where action-conditioned
attention is diverted from task-relevant regions to a localized patch. To demonstrate the threat, we propose
Attention-Guided Semantic Disruption (AGSD), an Expectation-over-Transformation (EOT) optimized printable patch that jointly (i) concentrates action-to-
vision attention on the patch and (ii) disrupts
vision-
language semantic
alignment, yielding strong cross-task and cross-architecture transfer. To mitigate such attacks, we introduce Structure-Aware Robust Fine-Tuning (SARF), a zero-inference-overhead defense that fine-tunes only the visual encoder using feature anchoring, policy-critical
attention correction, and
language-guided
geometric consistency restricted to semantically relevant regions. On LIBERO, SARF reduces OpenVLA's failure rate under AGSD from 100% to 14.2%-56.8% (28.6% average) across suites while preserving clean performance, and on a real PiPER manipulator it improves average success under AGSD from 23.0% to 65.0%. These results highlight mechanism-level robustness as a practical path to securing VLA
robots against physical
attention hijacking.
Read the full abstract ▾
All authors Jinquan Zhang, Dongfu Yin, Run Yang, Yufeng Yan, Zhen Tian, F. Richard Yu
Artificial Intelligence · Cross-listed from cs.LG
Agentic Reinforcement Learning with Self-Distilled Reward Shaping
Ranxu Zhang, Guinan Chen, Chenshaodong et al. (+5 more)
Agentic reinforcement learning enables LLM
agents to learn through interaction, but sparse trajectory-level rewards reveal success without identifying which intermediate decisions deserve credit. Training-only privileged skills can provide denser supervision by allowing the same frozen policy snapshot to rescore fixed
tokens from skill-free trajectories while conditioned on task-matched procedural skills. Existing methods, however, do not jointly calibrate teacher scores across interaction steps, relate teacher confidence to realized returns, and integrate the resulting signal into native reward-to-advantage construction. We introduce Agentic Reinforcement Learning with Self-Distilled Reward Shaping (ADRS), a framework for constructing return-associated
token-level credit for multi-turn
language agents. ADRS centers and normalizes privileged
token scores within each step, modulates them with a return-associated Teacher Value Advantage (TVA) gate based on within-group confidence–return association, and incorporates the gated
token signal into native RL credit construction. Together, these components determine what the teacher prefers, when that preference is return-relevant, and how it enters the native reinforcement-learning credit path, while keeping rollouts and inference skill-free. Finally, experiments across three interactive
benchmarks show that ADRS consistently improves performance on long-horizon tasks, with gains persisting across RL backbones, reduced-data settings, unseen tasks, and extended training. For anonymous review, our code is available at the following the link: https://github.com/gitrxh/ADRS-arxiv
Read the full abstract ▾
All authors Ranxu Zhang, Guinan Chen, Chenshaodong, Jinghao Lin, Xiaozhou Xu, Sunzhe, Yanyong Zhang, Chao Wang
Artificial Intelligence · Cross-listed from cs.SE
Fail-Fast, Restart-Smart: Early Failure Prediction and Restart for SWE Agentic Tasks
Chenyu Wang, Yunbo Lyu, Junda He et al. (+4 more)
Software engineering (SWE)
agents resolve repository-level issues through long trajectories that grow increasingly expensive as context accumulates. Failed runs tend to be longer and exhibit redundant exploration or looping, suggesting that some failures may be detectable before completion. Early termination, however, risks interrupting trajectories that would otherwise succeed; conversely, an unsuccessful trajectory may still contain useful repository edits. We present FailFast-RestartSmart, a two-stage controller for a single active trajectory. FailFast is a lightweight 0.6B monitor trained with terminal and dense fail-to-pass supervision to predict failure from observable prefixes without policy logits or hidden states. Upon an alarm, RestartSmart launches a fresh same-policy rollout without prior
prompt history and offers the interrupted repository diff as an optional overlay that the
agent may inspect, apply, or discard. On SWE-bench Verified, a monitor trained solely on Qwen3.6-27B trajectories transfers to three other policies, including a closed-API model, and saves 14.6%-20.4% of execution
tokens at a target 5% false-positive rate; on Qwen3.6-27B, its 20.4% saving exceeds the 12.5% achieved by our per-step AgentStop adaptation. At a target 25% false-positive rate, RestartSmart raises Qwen3.6-27B resolution from 66.6% to 71.8%, whereas cold restart reaches only 66.8%. Together, these results support early termination with sequential same-policy recovery.
Read the full abstract ▾
All authors Chenyu Wang, Yunbo Lyu, Junda He, Zhou Yang, Chenxing Zhong, Yaniv Harel, David Lo