Tech & Business

NEWSWIRE

LIVE

Hidden-State Probes Predict LLM Agent Failure Early, Saving Up to 47% Inference Compute

Researchers show that failure in LLM agent episodes is predictable early from the agent's internal representations. Lightweight per-round probes on hidden activations anticipate eventual episode failure as early as the first inter...

AI

o3-mini Achieves Higher Accuracy Than o1-mini Without Longer Reasoning Chains

A systematic analysis of reasoning chain length across o1-mini and o3-mini variants on the Omni-MATH benchmark finds that o3-mini (m) achieves superior accuracy without requiring longer reasoning chains than o1-mini. Accuracy gene...

AI

FirstResearch Framework Makes AI-Generated Scientific Questions Auditable

A new framework called FirstResearch introduces a structured Research Question Certificate to make the first research question proposed by LLM scientific discovery agents inspectable before downstream execution. The certificate re...

AI

RuBench Tests Coding Agents on Russian-Language Tasks From Real Repositories

A new benchmark called RuBench 1.0 evaluates product-grade coding agents on 25 repository-level tasks mined from recent fix commits in five live open-source repositories (aiohttp, aiogram, Laravel, NestJS, Fastify) spanning Python...

AI

HARC Couples Harmfulness and Refusal Directions for Stronger LLM Safety Alignment

Researchers introduce HARC (Harmfulness-And-Refusal Coupling), a fine-tuning method that pairs harmfulness and refusal directions across both prompt and response positions in the residual stream. The method achieves the strongest ...

AI

ROK-FORTRESS Benchmark Reveals Language and Geopolitics Shape LLM Safety Behavior

A new bilingual benchmark called ROK-FORTRESS shows that language and geopolitical context interact to shape large language model safety behavior in ways translation-only evaluations miss. Using the English-Korean language pair an...

PODCAST

Conversations & interviews

View All

An Exclusive Sit-Down Interview with President Michael Elliott - WAMH Radio

1h 0m

WAMH Radio · Interview

An Exclusive Sit-Down Interview with President Michael Elliott

Pierce Berke sits down with Michael A. Elliott, the 20th President of Amherst College, for an exclusive conversation about AI in liberal arts education, the future of higher education, and what it means to lead an institution through unprecedented technological change.