Alignment, interpretability, red-teaming, bias and privacy: the research on what these systems do when they misbehave.
Latent visual reasoning (LVR) inserts supervised latent tokens between perception and answer generation in vision-language models (VLMs). The field uses alignment between these latents and their visual targets, i.
Diffusion-based vision-language-action (VLA) models often inherit the image-generation view: actions are generated by iterative denoising. We argue that VLA action generation has a different condition-target structure: the policy is conditioned on rich observations, language, and state, but predicts only a compact, low-dimensional action chunk.
Recent Large Language Models (LLMs) have shown impressive reasoning abilities; but they are still susceptible to hallucinations, intermediate reasoning mistakes, and unreliable reasoning results in complex mathematical reasoning problems. In this study, we introduce a critic-based heterogeneous multi-agent approach to improve the dependability of mathematical reasoning.
Recent advancements in Vision-Language Models (VLMs) have significantly enhanced their ability to interpret complex visual semantics, yet their capacity for chronological reasoning remains under-explored. In this paper, we introduce a novel benchmark specifically designed to evaluate how VLMs perceive and reason about chronological information within and across images.
Agents increasingly generate SQL, orchestrate pipelines, and automate data analysis on behalf of users. While recent work improves query correctness, correctness is not safety.
arXiv:2606. 04366v1 Announce Type: new Abstract: Conventional patchified Transformers operate on uniform spatial partitions, distributing computational effort evenly across the domain irrespective of local features.
arXiv:2606. 04807v1 Announce Type: new Abstract: Mitigating social bias in Large Language Models (LLMs) presents a distinct alignment challenge: unlike verifiable tasks, bias lacks a single ground truth, creating a high-variance, subjective reward landscape.
arXiv:2606. 04280v1 Announce Type: cross Abstract: Contrastive learning has become a leading paradigm for self-supervised representation learning, yet the conditions under which it recovers meaningful latent geometry remain incompletely understood.
arXiv:2304. 10891v3 Announce Type: replace-cross Abstract: Transformer-based models are becoming a central paradigm in autonomous driving because they can capture long-range spatial dependencies, multi-agent interactions, and multimodal context across perception, prediction, and planning.
arXiv:2606. 04167v1 Announce Type: cross Abstract: We tackle the Metro Network Expansion Problem (MNEP), a subset of the Transport Network Design Problem (TNDP), which focuses on expanding metro systems to satisfy travel demand.
arXiv:2506. 10912v4 Announce Type: replace Abstract: Toxicity remains a leading cause of early-stage drug development failure.
arXiv:2601. 06196v3 Announce Type: replace-cross Abstract: Large language models (LLMs) frequently generate factually incorrect or unsupported content, commonly referred to as hallucinations.
arXiv:2509. 02655v3 Announce Type: replace-cross Abstract: Many AI alignment discussions of "runaway optimisation" focus on RL agents: unbounded utility maximisers that over-optimise a proxy objective (e.
arXiv:2510. 13704v2 Announce Type: replace-cross Abstract: Recent works have proposed accelerating the wall-clock training time of actor-critic methods via the use of large-scale environment parallelization; unfortunately, these can sometimes still require large number of environment interactions to achieve a desired level of performance.
arXiv:2510. 26219v3 Announce Type: replace-cross Abstract: Test-time alignment of large language models (LLMs) attracts attention because fine-tuning of LLMs requires high computational costs.
arXiv:2603. 10044v2 Announce Type: replace-cross Abstract: A safety score earned on a benchmark need not predict how the same model behaves once it is wrapped in an agentic scaffold the benchmark never tested.
arXiv:2606. 04227v1 Announce Type: cross Abstract: We present an algorithmic framework for incremental maintenance of first sheaf cohomology $H^1(X; \mathcal{F})$ on dynamically evolving 1-dimensional cellular complexes equipped with finite-dimensional cellular sheaves.
arXiv:2603. 23841v2 Announce Type: replace-cross Abstract: While Large Language Models (LLMs) are increasingly used as primary sources of information, their potential for political bias may impact their objectivity.
arXiv:2603. 28762v2 Announce Type: replace-cross Abstract: Modern Text-to-Image (T2I) diffusion models have achieved remarkable semantic alignment, yet they often suffer from a significant lack of variety, converging on a narrow set of visual solutions for any given prompt.
arXiv:2606. 04823v1 Announce Type: new Abstract: Large language models (LLMs) are fluent on open-ended tasks, yet in agentic settings, where a system must plan, use tools, and act over extended horizons, fluency does not ensure reliable delivery.