Prism: Dynamic Sparse Attention for Native 2K Joint Video-Audio Generation Model Training
Research
Research News
New papers, breakthroughs and academic AI research
RocketAgent: A Long-Horizon Engineering Agent for Multidisciplinary Design of Liquid-Rocket Thrust Chambers
Research
Judged Useless, Queried Anyway: Tool-Using Agents Rarely Turn Their Own Evidence Judgments into Stopping Decisions
Research
Process Constitutions and Process Stewards: Towards the Next Generation of BPM for Agentic Organizations
Research
Bridging the Evidence-to-Execution Gap:A Reflective Agent for Multi-Objective Peptide Design
Research
OntoInk: Interactive Ontology Visualization, Validation, and Reasoning
Research
Benchmarking Jailbreak Guardrails for Embodied Agents
Research
Copies or Sources? Measuring How LLM Aggregators Count Restated Evidence in Multi-Agent Systems
Research
Reward Stealing Attack on Large Language Models
Research
One Token Can Be Enough: Bridging Prompting and Activation Steering with Prefix Steering
Research
How Long, Not How Close: A Learned Temporal Metric for Planning in Latent World Models
Research
Teaching a Minimalist Machine to Discover Recursive Programs for Arithmetic
Research