Loading activity…
@dair_ai
Democratizing AI research, education, and technologies. Learn about AI Agents for FREE at https://academy.dair.ai/courses/elements-of-ai-agents
AI
Exo's creators say models can use runtime logs and modify and rebuild their harness while optimizing a task.
AI
RoboArchitect's researchers report average success rose from 9.7% to 43.3% on unseen scenes.
AI
Skill++'s authors report that it selected every required skill for 76.3% of queries, versus 36.4% with original metadata.
AI
LifeVal draws on a survey of 3,995 respondents to test AI on tasks involving caregiving, shopping, finances and government services.
XAI
Hongyan Chang and colleagues report higher detection than a token-level baseline in mathematical-reasoning tests using API access alone.
AI
DAIR.AI Academy says Clairic returns requested values as typed answers with confidence scores rather than generating text.
AI
Researchers at Meta and Carnegie Mellon describe a workflow for selecting tools from catalogs with more than 10,000 options.
AI
Google Cloud AI Research says all frontier models scored under 50% on its physical-engineering test.
AI
The researcher reports Drex Max got both labels right on 76.5% of 149 test papers, versus 74.5% for Jev.
AI
Google Cloud AI Research coauthors say OO-Harness builds specialized harnesses from execution records, then routes tasks among them.
AI
A post says the meta-agent reads task trajectories without success labels, then builds interfaces while the model stays fixed.
AI
The @dair_ai roundup highlights ten papers, including verified quantum-circuit designs and methods for managing agents’ evidence and memory.
AI
A post recounting a talk with founder Harrison Chase says the share fell over two months as token use rose.
AI
A post says it revisits retrieved reflections that guided a task, correcting outdated or overgeneralized lessons.
AI
dair_ai highlights research on editing agent harnesses, navigating document collections by entity and checking shared errors in agent runs.
AI
Omar Sar0 said Jev-as-a-Judge improves LLM judge reliability through consistency and is useful for judges, verifiers, and continuous monitoring. DAIR AI separately promoted it for agent evaluation, and Sar0 said he is working on a small benchmark to test that more broadly against other decision APIs.
AI
dair_ai highlights research on editable model context, self-organizing coding agents and how agents make decisions.
AI
A post about a Salesforce AI Research paper says Critical-State RL trains the tool call whose action changes the outcome, rather than spreading reward across the full sequence.
AI
A post about the SkillGym paper says it turns human-written skills into 2,756 training environments with code-based checkers, then fine-tunes a model on 8,364 successful trajectories.
AI
DAIR.AI’s selection highlights HySparse2’s reported cuts to prefill compute, GAVEL’s graph-based approach to robot planning and XYEval’s tests of misleading user hints.
