127,664Total entries

60Added by human

127,604Found by agent

13Categories

Knowledge catalogue

Search: “safety”

22,392 results

Safety

Long-Horizon RL Needs Muon

DGX agent

arXiv:2610.20256v1 Announce Type: new Abstract: Long-horizon agents must coordinate reasoning, tool use, and environment feedback over extended interactions. Learning from sparse outcome rewards requi

safetyarxiv-cs-lg

15 Oct 2026

Safety

Lost in Translation: Why Lower Bias Scores Don't Mean Fairer Models

DGX agent

arXiv:2610.20479v1 Announce Type: new Abstract: Bias audits of multilingual language models usually compare a model's average stereotype score in English with its average in another language and read

Search: “safety”

Long-Horizon RL Needs Muon

Lost in Translation: Why Lower Bias Scores Don't Mean Fairer Models

MAEM: Mitigating Recency Bias via a Replay-Free, Capacity-Controlled Dynamic Mixture of Experts for Class-Incremental Graph Learning

Maximum-Entropy Exploration with Future State-Action Visitation Measures

MemDuet: Learning Dual-Process Memory for Long-Context Reasoning

MIPS: Mitigating Mode Collapse in Reinforcement Learning for Large Outcome Spaces

MMLA-RSI: Recursive Self-Improvement through RTT and RTMA

Model What Matters: Decision-Guided Selective World Modeling for Language Agents

Multi-Sensitive Distributional Fairness: Joint Wasserstein Audits and Shapley Allocation

Neos: Learning to Interleave Search and Visual Reasoning

NePPO: Near-Potential Policy Optimization for General-Sum Multi-Agent Reinforcement Learning

Neural Supervised Clustering for Learning Individualized Treatment Rules with Many Treatments

Noisy-Label Calibration with Multiple Unreliable Labels

Not All Criteria Matter: Importance-Aware Criteria Weighting for Rubric-Based LLM Evaluation

Occupancy Modeling via Drifting: One-Step Generation Enables Direct Policy Extraction

Of Errors and Echoes: Responsibility-Aware Credit Assignment for Tool-Use Agents

On the Complexity Gap Between POMDPs and Multi-Environment POMDPs

On the Convergence of Semi-Gradient TD: A Mean Field Perspective

On the Limits of Representation-Based Test-Time Adaptation for Multimodal LLMs

One-shot Human Video-Guided Adaptation for Vision-Language-Action Models

One Task, Many Interfaces: Simple Cross-Interface Consistency Training Improves LLM Agent Robustness

OneLA: Learning Many Worlds from Latent Actions

Outcome-Routed Distillation and Reinforcement Learning for Diffusion Language Models

Outperforming 1000 Euclidean Layers With Hyperbolic Goal-Conditioned RL

PA1-OPD: One-Step Prefix-Augmented On-Policy Distillation for Language Model Pretraining

Paraphrase Consensus Is Not Self-Consistency: Instruction Reliance and Robustness in VLA Policies

PEARL: Process-Experience Aware Reinforcement Learning for Multi-Turn Agents

PhySGA: A Physics-Space Grounded Agentic Framework for Generalizable and Interactive 3D Composition

Physical Time as a Missing Dimension of Robot Policy Memory Dynamics

Physics-Shaped, Not Physical: Auditing Physics-Guided Underwater Image Restoration

PolyMorph: Learning Locomotion Policy for Cross-Embodiment Adaptation

PopGen: Generating Semantically Diverse Populations for Zero-Shot Coordination

Post-Training, Finetuning, and Steering:A Field Guide for Curating Training Data

Pre-Trained Language Models as Zero-Shot Tools for Social Science Research

RAP-WAM: Learning World-Action Models through Predictive Rehearsal and Action Perturbations

ReAl: Refine neural operators via aligning dynamical flow

Recovering Reliable Supervision via Critique-Guided On-Policy Distillation

Refine Actions If Needed: Adaptive Action Chunk Refinement for Robot Manipulation

Reinforcement Learning via Advantage Guidance

Relational Semantic Grounding for Test-Time Discovery

Rent-Seeking Language Models: Reward Models Set the Prices, Frequent Audits Curb the Rent

Representation Conditioning for Generation

RetailAgent: Structured Adverse Timing in Self-Conditioned Multimodal LLM Trading Agents

Retell, Reward, Repeat: Reinforcement Learning for Narrative Theory-Informed Story Retelling

RL Post-Training Diffusion Language Models Along Their Own Generation Paths

RouteWeaver: Weaving Mode Selection and Execution into Unified LLM Routing

RTAformer: Relation-Transition-Aware Transformer for Video Scene Graph Generation

Rule-based and reinforcement-learning policies for resource allocation in a pediatric echo lab: a simulation study