All entries

Verifiable Benchmarking of Long-Horizon Spatial Biology

arXiv:2605.28065v1 Announce Type: new Abstract: AI agents are increasingly useful for biological data analysis, but existing benchmarks mostly test broad biological knowledge, executable workflows, or

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Verified Misguidance: Measuring Structural Citation Failures in Search-Augmented LLMs

DGX agent

arXiv:2605.28565v1 Announce Type: cross Abstract: Users of search-augmented LLMs rely on citations as evidence that responses are grounded in real sources, and rarely verify the cited pages themselves

safetyarxiv-cs-ai

28 May 2026

VeriTrip: A Verifiable Benchmark for Travel Planning Agents over Unstructured Web Corpora

arXiv:2605.28683v1 Announce Type: new Abstract: Existing benchmarks have laid the foundation for travel planning agents by establishing API-centric paradigms. However, as the capabilities of Autonomou

VibeSearchBench: Benchmarking Long-horizon Proactive Search in the Wild

arXiv:2605.27882v1 Announce Type: cross Abstract: LLM-based agents score well on search benchmarks, yet real users consistently find results unsatisfying, revealing a persistent evaluation-experience

ViCA: Efficient Multimodal LLMs with Vision-Only Cross-Attention

arXiv:2602.07574v2 Announce Type: replace-cross Abstract: Modern multimodal large language models (MLLMs) adopt a unified self-attention design that processes visual and textual tokens at every Transf

VideoCanvas: Unified Video Completion from Arbitrary Spatiotemporal Patches via In-Context Conditioning

arXiv:2510.08555v2 Announce Type: replace Abstract: Existing controllable video generation methods are typically designed for rigid, task-specific settings, such as first-frame image-to-video, inpaint

VidPrism: Heterogeneous Mixture of Experts for Image-to-Video Transfer

arXiv:2605.28229v1 Announce Type: cross Abstract: With the rapid development of pre-training technologies, adapting large-scale Vision-Language Models (VLMs) for video understanding ie image-to-video

researcharxiv-cs-ai

Visualizing Latent Phase Structures in Locomotion Policies: A Multi-Environment Study with Temporal Feature Extension

arXiv:2605.28186v1 Announce Type: cross Abstract: Deep reinforcement learning (DRL) has been shown to achieve high performance on locomotion control tasks in MuJoCo benchmarks such as HalfCheetah, Ant

VITAL: Visual-Semantic Dual Supervision for Enhanced and Interpretable Latent Reasoning in Medical MLLMs

arXiv:2605.28422v1 Announce Type: cross Abstract: Latent reasoning enables reasoning over continuous hidden states rather than explicit tokens, avoiding the language bottleneck and inference overhead

VLA-Hijack: A Transferable Patch Attack against Vision-Language-Action Models via Visual Proprioception Hijacking

arXiv:2605.28083v1 Announce Type: new Abstract: While Vision-Language-Action (VLA) models have emerged as powerful generalist policies, their severe vulnerability to adversarial patches significantly

safetyarxiv-cs-cv

VLM-Based Advanced Rider Assistance System for Motorcycle Safety

arXiv:2605.27948v1 Announce Type: new Abstract: Motorcycles face disproportionately high crash risks compared to cars due to limited protection and heightened sensitivity to surface hazards, yet Advan

safetyarxiv-cs-ro

VLMs May Not Globally Enhance Human Alignment over LLMs During Natural Reading

arXiv:2605.28818v1 Announce Type: new Abstract: Large language models (LLMs) have become increasingly useful computational models of human language processing, but it remains unclear whether vision-la

safetyarxiv-cs-cl

Voluntary Collusion with Secret Tools in Competing LLM Agents

arXiv:2605.27593v1 Announce Type: new Abstract: Even when a tool is explicitly described as unfair and harmful to others, ostensibly safety-aligned LLM agents still voluntarily engage in secret collus

VULPO: Context-Aware Vulnerability Detection via On-Policy LLM Optimization

arXiv:2511.11896v3 Announce Type: replace-cross Abstract: Large language models (LLMs) have recently shown strong potential in vulnerability detection (VD). However, accurately detecting vulnerabiliti

Weak Convergence Analysis of Online Neural Actor-Critic Algorithms

model-releasesarxiv-cs-lg

arXiv:2403.16825v2 Announce Type: replace Abstract: We prove that a single-layer neural network trained with the online actor critic algorithm converges in distribution to a random ordinary differenti

WeatherCity: Urban Scene Reconstruction with Controllable Multi-Weather Transformation

arXiv:2602.22096v2 Announce Type: replace Abstract: Editable high-fidelity 4D scenes are crucial for autonomous driving, as they can be applied to end-to-end training and closed-loop simulation. Howev

What Are We Measuring in NLG? A Meta-Analysis of Evaluation Trends 2020-2025

arXiv:2601.07648v2 Announce Type: replace Abstract: As Natural Language Generation (NLG) dominates modern NLP, scalable evaluation remains a critical bottleneck. Consequently, LLM-as-a-judge (LaaJ) ad

safetyarxiv-cs-cl

What Frozen VLAs Already Know About Success: A Probing Study of Value-Like Structure in Foundation Robot Policies

arXiv:2605.28527v1 Announce Type: new Abstract: Vision--language--action (VLA) policies are trained to imitate actions; their loss never asks them to estimate reward, progress, or future success. Thei

safetyarxiv-cs-ro

What-If World: A Causal Benchmark for General World Models in Embodied Scenarios

arXiv:2605.27589v1 Announce Type: new Abstract: Video generation models are increasingly used as world simulators for tasks like driving and robotic manipulation. What matters in these settings is not

When Confidence Misleads: Suffix Anchoring and Anchor-Proximity Confidence Modulation for Diffusion Language Models

arXiv:2605.28181v1 Announce Type: new Abstract: Diffusion language models decode text by iteratively denoising masked token sequences, making the choice of which positions to decode a central inferenc

When Context Flips, Safety Breaks: Diagnosing Brittle Safety in Aligned Language Models

arXiv:2605.27851v1 Announce Type: new Abstract: Safety benchmark scores provide incomplete evidence of deployment readiness: aligned language models often adhere to rigid rules even when a situational

When Discourse Pressures Conflict: Information Structure in Vision-Language Model Outputs

arXiv:2605.28346v1 Announce Type: new Abstract: Vision-language models (VLMs) are increasingly evaluated for whether they identify the right visual content, but little is known about whether they expr

When do complex-valued neural networks help? A study of representation, geometry, and optimization

model-releasesarxiv-cs-lg

arXiv:2605.27673v1 Announce Type: new Abstract: Complex-valued Neural Networks (CVNNs) are often motivated by domains where information is naturally encoded in magnitude and phase. Yet complex-valued

Agents

When Does Memory Help Multi-Trajectory Inference for Tool-Use LLM Agents?

arXiv:2605.28224v1 Announce Type: new Abstract: Multi-trajectory inference for tool-use LLM agents - generating multiple reasoning attempts and selecting among them - benefits from transferring knowle

agentsarxiv-cs-ai

When Helpful Context Leaks: Privacy Risks in Domain-Adapted ASR

arXiv:2605.28211v1 Announce Type: new Abstract: SpeechLLMs are increasingly deployed in professional settings where domain customisation is standard practice: users supply context in prompts with sens

When Interpretability Is Unequally Distributed: Fairness in Hybrid Interpretable Models

model-releasesarxiv-cs-lg

arXiv:2605.28626v1 Announce Type: new Abstract: Hybrid interpretable models combine a transparent component with a black-box model by assigning some examples to the former and deferring the rest to th

Local Ai

When NPUs Are Not Always Faster: A Stage-Level Analysis of Mobile LLM Inference

arXiv:2605.27435v1 Announce Type: cross Abstract: Deploying large language models (LLMs) on mobile devices increasingly relies on heterogeneous execution, yet no prior study has systematically charact

local-aiarxiv-cs-ai

When pre-training hurts LoRA fine-tuning: a dynamical analysis via single-index models

arXiv:2602.02855v2 Announce Type: replace Abstract: Pre-training on a source task is usually expected to facilitate fine-tuning on similar downstream problems. In this work, we mathematically show tha

safetyarxiv-cs-lg

When prompt perturbations break your A/B test: A valid statistical test for generative surveying

arXiv:2605.27463v1 Announce Type: cross Abstract: Generative surveying -- where collections of LLM-based personas provide feedback on messages -- has emerged as a cheap and scalable alternative to tra

researcharxiv-cs-ai

When Seekers Are Hard to Help: Evaluating Emotional Support Dialogue Systems in Worst-Case Interactions

arXiv:2605.28228v1 Announce Type: new Abstract: Emotional Support Dialogue Systems (ESDSes) are increasingly evaluated and trained with LLM-simulated seekers. However, such simulated seekers often beh

When Think-with-Image Meets Safety: What Determines Multimodal Jailbreak Robustness?

arXiv:2605.27932v1 Announce Type: cross Abstract: Think-with-image reasoning is emerging as a new inference paradigm for large vision-language models, but its safety implications remain poorly underst

Local Ai

Where Does Toxicity Live? Mechanistic Localization and Targeted Suppression in Language Models

arXiv:2605.27997v1 Announce Type: cross Abstract: Large language models frequently generate toxic, hateful, or harmful content, yet existing mitigation methods rely on costly retraining or output-leve

local-aiarxiv-cs-ai

Where LLM Annotators Fail: Label-Free Learning on Graphs with LLMs

arXiv:2605.27913v1 Announce Type: new Abstract: Node classification on graphs often requires labeled nodes, yet obtaining labels at graph scale is expensive. When node attributes contain semantic cont

researcharxiv-cs-lg

Where Rollouts Begin: Low-Load, High-Leverage First-Token Diversification for RLVR

arXiv:2605.28295v1 Announce Type: new Abstract: Reinforcement Learning with Verifiable Rewards (RLVR) trains reasoning models without labeled trajectories, relying on grouped rollouts to expose the po

Which Heads Matter for Reasoning? RL-Guided KV Cache Compression

arXiv:2510.08525v3 Announce Type: replace Abstract: Reasoning large language models exhibit complex reasoning behaviors via extended chain-of-thought generation that are highly fragile to information

Tutorials

Which Pretraining Paradigm Better Serves Spatial Intelligence? An Empirical Comparison of Vision-Language and Video Generation Models

arXiv:2605.28132v1 Announce Type: new Abstract: Spatial intelligence requires visual representations that capture both semantic objects and geometric structure in the physical world. To support this,

tutorialsarxiv-cs-cv

Who Uses AI? Platform Selection and the Measurement of Occupational AI Exposure

arXiv:2605.21743v2 Announce Type: replace Abstract: Conversation logs from AI platforms are increasingly used to measure occupational exposure to artificial intelligence, but the users observed in the

Whose Is This?: Context-Aware Object Ownership Inference with Uncertainty-Guided Questioning

arXiv:2605.28087v1 Announce Type: new Abstract: Service robots must infer object ownership to correctly interpret instructions such as 'bring me my cup.' However, ownership is a latent attribute that

researcharxiv-cs-ro

Whose Name Comes Up? III: Persona Prompting Effects in LLM-Based Scholar Recommendation

arXiv:2605.28187v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly used as scholar recommenders, shaping who is seen as an expert in academia. Existing audits remain Engli

Tutorials

Why Gaussian Diffusion Models Fail on Discrete Data and How to Prevent It?

arXiv:2604.02028v2 Announce Type: replace Abstract: Diffusion models have become a standard approach for generative modeling in continuous domains, yet their application to discrete data remains chall

tutorialsarxiv-cs-cl

Why LLMs Fail at Causal Discovery and How Interventional Agents Escape

arXiv:2605.27567v1 Announce Type: new Abstract: Causal discovery is a cornerstone of scientific reasoning, yet whether large language models can perform it reliably remains an open question. Recent be

Why We Need Speech to Evaluate Speech Translation

arXiv:2605.28227v1 Announce Type: new Abstract: Speech translation models are increasingly capable of preserving speech-specific information (e.g., speaker gender, prosody, and emphasis), yet evaluati

Worker Disagreement Reveals Sharp Directions in Local SGD

arXiv:2605.27739v1 Announce Type: cross Abstract: Deep neural network training often exhibits highly anisotropic loss geometry, where a few sharp dominant Hessian directions coexist with a large flatt

researcharxiv-cs-ai

xKV: Cross-Layer KV-Cache Compression via Aligned Singular Vector Extraction

arXiv:2503.18893v2 Announce Type: replace Abstract: Long-context Large Language Models (LLMs) enable powerful applications but incur high memory costs due to the key-value states (KV-Cache). Recent st

safetyarxiv-cs-cl

XTransfer: Modality-Agnostic Few-Shot Model Transfer for Human Sensing at the Edge

arXiv:2506.22726v4 Announce Type: replace Abstract: Deep learning for human sensing on edge systems presents significant potential for smart applications. However, its training and development are hin

Applications

You Are in Control of Your State: Why Human Outcomes Are Controllable Through Causal State Intervention

arXiv:2605.27580v1 Announce Type: new Abstract: A central puzzle for the behavioural sciences and for human-facing artificial intelligence is the persistence of within-person variability. The same ind

applicationsarxiv-cs-ai

You Live More Than Once: Towards Hierarchical Skill Meta-Evolving

arXiv:2605.28390v1 Announce Type: new Abstract: Test-time skill evolving is regarded as a new paradigm for enhancing deployed agentic systems. Existing works mainly focus on hard-coded skill evolving

You Only Align Once: Propagating Cooperative Behaviors in Multi-Agent Systems through Seed Agents

model-releasesarxiv-cs-cl

arXiv:2605.27586v1 Announce Type: cross Abstract: Ensuring agent behaviors in distributed open multi-agent systems remains challenging, especially as populations grow and unaligned agents may exist. W