arXiv:2602.01740v3 Announce Type: replace Abstract: Video language models (Video-LLMs) are prone to hallucinations, generating plausible but ungrounded content when visual evidence is weak, ambiguous, or biased. Existing methods, such as contrastive decoding (CD), rely on random perturbations to construct contrastive data for hallucination mitigation, but often fail to target the visual cues that drive hallucination […]
A Mechanism-Coupled Split Window Network for Medium- to High-Resolution Land Surface Temperature Retrieval
arXiv:2509.04991v2 Announce Type: replace-cross Abstract: Land surface temperature (LST) is a fundamental physical variable in land-atmosphere interactions, surface energy budgets, and climate processes. LST derived from medium- to high-resolution thermal infrared (TIR) observations effectively reveals thermal environmental disparities across distinct landscape units. However, achieving accurate, robust, and globally generalizable LST retrieval remains challenging under complex […]
SlimSearcher: Training Efficiency-Aware Web Agents via Adaptive Reward Gating
arXiv:2606.07074v1 Announce Type: cross Abstract: Deep research agents have demonstrated remarkable capabilities in complex information-seeking tasks, yet this power comes at a steep computational cost. Driven by accuracy-focused training paradigms, current models adopt brute-force strategies characterized by blind tool dependency and performative reasoning-generating long, redundant trajectories that are far from necessary for resolving these tasks, […]
Position: Don’t Just “Fix it in Post”: A Science of AI Must Study Training Dynamics
arXiv:2606.06533v1 Announce Type: new Abstract: What would it mean to have a scientific understanding of AI? Models are not static objects: they are snapshots of time-evolving processes shaped by data, objectives, architectures, and optimization dynamics. Yet much of AI research treats models as fixed artifacts, analyzing behaviors after training rather than asking why they emerge. […]
The Three-Ring Architecture: Governing Agents in the Era of On-Platform Organisations
arXiv:2606.07119v1 Announce Type: cross Abstract: The current phase of enterprise AI deployment faces a structural failure: organisations are acquiring agentic capability without the infrastructure to govern it. The result is expected to reproduce the error of the first wave of AI deployment: decentralised intelligence without a federation layer leading to a 95% project failure rate. […]
CrowdMath: A Dataset of Crowdsourced Mathematical Research Discussions
arXiv:2606.06526v1 Announce Type: new Abstract: Large language models have made substantial progress on mathematical reasoning, but existing benchmarks typically evaluate well-specified problems with final answers, step-by-step solutions, or complete proofs. They do not capture collaborative open-problem solving: a setting in which participants propose partial arguments, identify gaps or errors in prior steps, repair flawed reasoning, […]
SV-Detect: AI-generated Text Detection with Steering Vectors
arXiv:2606.07313v1 Announce Type: cross Abstract: Detecting machine-generated text is especially difficult under distribution shift, such as transfer across domains, source models, and editing attacks. We propose a fake-text detector based on steering vectors extracted from the hidden representations of a frozen language model. At each layer, we construct a direction that separates human-written from machine-generated […]
Attack Selection in Agentic AI Control Evaluations Meaningfully Decreases Safety
arXiv:2606.06529v1 Announce Type: new Abstract: An attacker that strategically chooses when to attack is much harder to catch than one that attacks indiscriminately. AI control is a safety framework for deploying capable but untrusted AI agents under the oversight of a weaker, trusted monitor and a limited human audit budget. Control evaluations stress-test these protocols […]
Multilingual Multi-Speaker Unit Vocoders: A Systematic Analysis of Discrete Speech Representations
arXiv:2606.06740v1 Announce Type: cross Abstract: Discrete speech units obtained via k-means clustering of self supervised embeddings entangle phonetic, speaker, and language information, causing speaker mixing and cross-lingual interference in multilingual multi-speaker speech generation. Despite growing use in Audio LLMs and speech to speech systems, unit vocoders remain underexplored. We analyze a BigVGAN based unit vocoder, […]
Lean4Agent: Formal Modeling and Verification for Agent Workflow and Trajectory
arXiv:2606.06523v1 Announce Type: new Abstract: Equipping Large Language Models (LLMs) to execute reliable multi-step workflows has become a central challenge in artificial intelligence. Despite recent advances in LLMs’ agentic capabilities, most agent systems still lack formal methods for specifying, verifying, and debugging their workflow and execution trajectories. This challenge mirrors a long-standing problem in mathematics, […]
Exploring Reinforcement Learning for Fluid Transitions Between Clinical Mental Healthcare and Everyday Wellness Support
arXiv:2606.06800v1 Announce Type: cross Abstract: Mental health struggles wax and wane, yet clinical and wellness interventions typically operate separately, causing frequent breakdowns at care transitions. We explore reinforcement learning (RL) as a means to build digital health systems that deliver clinical and wellness interventions proactively, as part of a coherent care journey. We ask: what […]
SafeGene: Reusable Adapters for Transferable Safety Alignment
arXiv:2606.06519v1 Announce Type: new Abstract: Open-weight LLMs are increasingly fine-tuned into customized assistants, but downstream fine-tuning can weaken safety alignment and make models more vulnerable to malicious prompts, even when the training data is not intentionally harmful. This creates a recurring safety recovery problem as target models are repeatedly updated with new task data or […]