Search: “safety”

Reliability Scales Inversely: Bigger Language Models Compound Mistakes Faster

arXiv:2607.18292v2 Announce Type: replace-cross Abstract: As language models scale, answers start truer but degrade faster: scaling buys capability but erodes reliability. The knowledge-gap account --

Self-Guided Process Reward Optimization with Redefined Step-wise Advantage for Process Reinforcement Learning

DGX agent

arXiv:2507.01551v3 Announce Type: replace-cross Abstract: Process Reinforcement Learning~(PRL) has demonstrated considerable potential in enhancing the reasoning capabilities of Large Language Models~

safetyarxiv-cs-cl

27 Jul 2026

Spectral Prior for Reducing Exposure Bias in Diffusion Models

arXiv:2607.22091v1 Announce Type: new Abstract: Diffusion models typically suffer from error accumulation during iterative sampling, commonly referred to as exposure bias. We reveal systematic frequen

Stocks should be ripping today given the move in oil. But they're not thanks to NVDA and it's unbelievable 250 Billion roundtrip with Open…

Stocks should be ripping today given the move in oil. But they're not thanks to NVDA and it's unbelievable 250 Billion roundtrip with OpenAI. There's no hiding anymore that the whole AI bubble is real

TextSLIP: Text Self-Supervised CLIP for Medical Report Generation

arXiv:2607.21970v1 Announce Type: new Abstract: Automating radiology report generation is important for improving reporting consistency and clinical workflows . While Contrastive Language--Image Pretr

The Coordination Gap: Multi-Agent Alternation Metrics for Temporal Fairness in Repeated Games

arXiv:2603.05789v5 Announce Type: replace-cross Abstract: Repeated multi-agent interactions require evaluation metrics that capture not only payoff distributions but also their temporal organization.

TRACE-ROUTER: Task-Consistent and Adaptive Online Routing for Agentic AI

arXiv:2607.22465v1 Announce Type: cross Abstract: Routing to select large language models (LLMs) with different cost-quality trade-offs has become a fundamental deployment feature of enterprise AI. Ex

Twins: Learn to Predict Unified Representations with Focal Loss

arXiv:2607.22531v1 Announce Type: new Abstract: Unified multimodal models seek a shared visual token space that supports both multimodal understanding and image generation. Discrete methods unify the

Very cool paper from Microsoft. The idea is to train agents on replayed teacher trajectories instead of live environment rollouts. On-policy…

Very cool paper from Microsoft. The idea is to train agents on replayed teacher trajectories instead of live environment rollouts. On-policy distillation for agentic tasks is expensive because every u

safetydair-ai--x

ViTacWorld: Scaling Visuo-Tactile World Models for Contact-Rich Robot Manipulation

arXiv:2607.22530v1 Announce Type: new Abstract: Contact-rich robot manipulation requires physical interaction cues that are often invisible to cameras, making tactile sensing essential for robust cont

safetyarxiv-cs-ro

Model Releases

WHBench: Evaluating Frontier LLMs with Expert-in-the-Loop Validation on Women's Health Topics

model-releasesarxiv-cs-cl

arXiv:2604.00024v2 Announce Type: replace Abstract: Large language models are increasingly used for medical guidance, but women's health remains under-evaluated in benchmark design. We present the Wom

When Ethics and Payoffs Diverge: LLM Agents in Morally Charged Social Dilemmas

arXiv:2505.19212v2 Announce Type: replace Abstract: Recent advances in LLMs have enabled their use in complex agentic roles, involving decision-making with humans or other agents, making ethical align

Why Large Language Models and Humans Converge and Diverge in Evaluating Creativity

arXiv:2607.22218v1 Announce Type: new Abstract: Despite the growing use of large language models (LLMs) as creativity evaluators, evidence of their alignment with human evaluations remains mixed, rais

important thread; don’t read just the first tweet (which has a caveat in the second).

important thread; don’t read just the first tweet (which has a caveat in the second). In case you think the problem of science slop is hypothetical: Here's the president of OpenAI retweeting wrong sci

26 Jul 2026

training to benchmarks ≠ getting to AGI

training to benchmarks ≠ getting to AGI This suggests that Opus 5 's gain on ARC-AGI-3 was the result of specific training to improve on that eval, and not a generalized increase in abstract reasoning

26 Jul 2026

All anyone has to do is go back 15 years and look at all the ridiculous predictions Elmo has made about Mars, self driving cars, his ridicul…

All anyone has to do is go back 15 years and look at all the ridiculous predictions Elmo has made about Mars, self driving cars, his ridiculous Boring Company. No one should take anything he says seri

I am surprised how few people are aware that the reasoning for OpenAI/Anthropic models is all encrypted. The 'reasoning' you see in the UI i…

OpenAI and Anthropic’s language models keep their internal reasoning encrypted; what users see in the UI is only a filtered summary of that reasoning. This practice was highlighted in a tweet by Sarah

The two giant generative AI startups were treated like gods for a couple years. Both are now facing massive pushback.

The two giant generative AI startups were treated like gods for a couple years. Both are now facing massive pushback. Anthropic employees go nuclear on the popularity of open source AI. It is getting

When a normal company does a bad thing, they take responsibility + apologize + outline how it won't happen again. OpenAI instead is like 'we…

When a normal company does a bad thing, they take responsibility + apologize + outline how it won't happen again. OpenAI instead is like 'we are entering a new era. this will happen again. no one amon

A Framework for Reputation Aware Uninorm-driven Consensus Algorithms for Blockchain Networks

arXiv:2607.20700v1 Announce Type: cross Abstract: The operation of blockchain is governed by consensus algorithms (CA). Several consensus mechanisms require significant computational power, while othe

A Self-Evolving Default Action for Cooperative Tasks with Continuous Action Space

arXiv:2607.18597v2 Announce Type: replace Abstract: Counterfactual credit assignment has proven effective in multi-agent reinforcement learning (MARL) for discrete action spaces, yet its extension to

A Unified Moral-Value Dataset for Instruction Tuning

arXiv:2607.21279v1 Announce Type: new Abstract: Large language models (LLMs) have developed rapidly and become valuable tools in everyday life. However, how to align LLMs to a particular set of human

Adaptive Depth Sparse Framework: Similarity-Driven Resource Allocation for Pre-Trained LLMs

arXiv:2607.21291v1 Announce Type: new Abstract: Large language models (LLMs) achieve strong generation and reasoning performance, but the Transformer architecture incurs high inference cost. Existing

Anticipate Before Acting: Future-State-Conditioned Vision-Language Navigation

arXiv:2607.18042v2 Announce Type: replace-cross Abstract: End-to-end vision-language navigation (VLN) with causal vision-language models maps instructions and egocentric observations directly to actio

Approximate Quantum State Preparation Through Proximal Policy Optimization

arXiv:2607.21121v1 Announce Type: cross Abstract: In this work, a quantum architecture search framework for approximate quantum state preparation (QSP) is proposed. QSP is a challenging task, since th

ARCO: Adaptive Rubrics with Co-Evolution for Multi-Step LLM-Based Agents

arXiv:2606.21262v2 Announce Type: replace Abstract: Reinforcement learning for multi-step LLM agents often relies on scalar rewards that indicate success but cannot explain why a trajectory is good or

ASTRA-Net: Anatomy-Specific Transfer and Representation Alignment for Drug-Induced Sleep Endoscopy Segmentation

arXiv:2607.21370v1 Announce Type: new Abstract: Quantitative drug-induced sleep endoscopy (DISE) requires reliable airway boundaries at specific anatomical levels. Pixel-level DISE annotations are sca

AttriMem: Attribution-Guided Process Feedback for Agent Memory Learning

arXiv:2607.21106v1 Announce Type: new Abstract: Effective memory is crucial for LLM agents, yet constructing it effectively remains challenging. A memory-construction policy decides what information t

Axolotl3D: a Unified Framework for Faithful 3D Shape Completion

arXiv:2607.20660v1 Announce Type: new Abstract: Recent 3D generative models produce high-quality geometry from a single image using large-scale priors and diffusion architectures. However, they assume

Belief Propagation in LLM World Models: Measuring Strategic Information Bias with Prediction Markets

arXiv:2607.20441v1 Announce Type: new Abstract: Every information ecosystem produces beliefs that shape strategic decisions. Both human analysts and AI systems inherit the blind spots of their informa

Beyond Sycophancy: Structured Resistance and Compliance in LLM Moral Reasoning

arXiv:2607.21558v1 Announce Type: new Abstract: Building socially calibrated large language models, which can learn from others without simply yielding to them, requires more than reducing sycophancy

Climate-resilient electric vehicle charging infrastructure for sustainable cities: An interpretable causal-ensemble framework for preventive maintenance and low-carbon mobility

arXiv:2607.21444v1 Announce Type: cross Abstract: Reliable electric vehicle (EV) charging infrastructure is a cornerstone of sustainable, low-carbon cities, yet urban climate stress such as extreme he

Confidently Deceptive: How Confidence Amplifies the Risk of LLM Deception

arXiv:2607.20444v1 Announce Type: cross Abstract: Large language models (LLMs) can produce deceptive responses: outputs that mislead users in service of a contextually or experimentally induced goal.

CSPF: A Constrained Shared-Private Fusion Method for Non-Verifiable Preference Evaluation

arXiv:2607.20862v1 Announce Type: new Abstract: At present, reliable evaluation of non-verifiable tasks remains challenging. Existing approaches often fail to adequately capture the diverse evaluative

DINOde: Continuous Vision-Text Alignment for Open-Vocabulary Semantic Segmentation

arXiv:2607.21371v1 Announce Type: cross Abstract: Open-vocabulary semantic segmentation (OVSS) leverages textual semantics to segment objects beyond predefined categories. While the self-supervised mo

Distribution-Alignment Bridge for Uncertainty-Aware Text-to-Video Retrieval

arXiv:2607.20984v1 Announce Type: new Abstract: This paper proposes the Distribution-Alignment Bridge (DAB), a framework that reconceptualizes text-to-video retrieval as a distribution alignment task

DTIF: Robust Loop Closure Detection via Delaunay Triangle Topology in Complex Forests

arXiv:2607.21138v1 Announce Type: new Abstract: Accurate forest inventory and large-scale mapping are essential for ecosystem monitoring and sustainable forest management. Multiple low-cost edge platf

DynaMark: A Reinforcement Learning Framework for Dynamic Watermarking in Industrial Machine Tool Controllers

arXiv:2508.21797v2 Announce Type: replace-cross Abstract: Industry 4.0's highly networked Machine Tool Controllers (MTCs) are prime targets for replay attacks that use outdated sensor data to manipula

EmoAgent-R1: Towards Multimodal Emotion Understanding with Reinforcement Learning-based Dynamic Agent Specialization

arXiv:2607.21013v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) have achieved impressive performance in multimodal emotion recognition (MER) tasks and lifted MER to a new leve

EmoSpace: Immersive Affective Image Generation Guided by Fine-Grained Emotion Prototypes

arXiv:2602.11658v2 Announce Type: replace Abstract: Immersive affective content generation aims to create visually compelling VR imagery with controllable emotional nuance, yet existing methods typica

Enhancing Explainable Cardiac Diagnosis with Guide-Grounded Multimodal LLMs

arXiv:2607.20814v1 Announce Type: new Abstract: The electrocardiogram (ECG) is a cornerstone of cardiac as- sessment, yet clinical deployment of deep learning models remains con- strained by limited i

Environment-Aware Channel Inference via Cross-Modal Flow: From Multimodal Sensing to Wireless Channels

arXiv:2512.04966v2 Announce Type: replace-cross Abstract: Accurate channel state information (CSI) underpins reliable and efficient wireless communication. However, acquiring CSI via pilot estimation

Evaluating Risks in Weak-to-Strong Alignment: A Bias-Variance Perspective

arXiv:2604.25077v2 Announce Type: replace Abstract: Weak-to-strong alignment offers a promising route to scalable supervision, but it can fail when a strong model becomes confidently wrong on examples

EvoSQL: Memory-Augmented Critic-Generator Co-Evolution for Text-to-SQL

arXiv:2607.20489v1 Announce Type: new Abstract: Text-to-SQL has advanced rapidly with large language models, but complex database queries still require reasoning beyond one-shot generation, including

Expert Behavior Prior Reinforcement Learning

arXiv:2607.21302v1 Announce Type: new Abstract: Behavior prior reinforcement learning (BPRL) has emerged as a promising paradigm to improve sample efficiency in online reinforcement learning (RL) by l

Explainability Framework for Policy-Aware Autonomous Agents

arXiv:2607.21209v1 Announce Type: cross Abstract: In the field of Artificial Intelligence, an agent is a system which is able to autonomously make decisions in order to reach a desired goal. As these

Explainable graph attention network for stress recognition (StressGAT) via differential action units

arXiv:2607.20819v1 Announce Type: new Abstract: Stress is a dynamic process characterized by significant individual variability in facial expression. Traditional architectures, such as Recurrent Neura

FELT: Generating Tactile Signals from Vision for Visuo-Tactile Manipulation

arXiv:2607.20683v1 Announce Type: new Abstract: The sense of touch is central to manipulation, especially when vision is occluded or ambiguous. Although combining vision and touch improves manipulatio

safetyarxiv-cs-ro