Evolutionary search with large language models (LLMs) can stall when progress requires ext…
Automating the search for effective harnesses is an important step toward enabling agents …
As interactive 3D worlds are increasingly used to study intelligent behavior, it becomes i…
We present Cogentic, a multi-agent harness for automated proof discovery on open research …
Recent autonomous machine learning engineering (MLE) agents have made significant progress…
Web text makes up the majority of pretraining data and is increasingly AI-generated. After…
Unlearning a fact in one language does not guarantee its removal in others as changing the…
On-policy distillation (OPD) is a promising approach for training language agents, providi…
Multi-Agent Path Finding (MAPF) aims to find collision-free paths for multiple agents in a…
Converting free-text radiology reports into structured labels supports cohort building, qu…
Speech-to-speech systems must solve tasks whose requirements emerge across conversational …
Language agents are expected to solve increasingly complex tasks, creating a growing need …
Reinforcement learning with verifiable rewards (RLVR) has become an important approach for…
An unsuccessful LLM agent rollout contains more information than its final reward: the obs…
Particle transport under multiple scattering is central to radiative transfer and plasma p…
Causal action verifiers gate an agent's state-changing tool calls by checking whether each…
Background. A component replacement in a language-model agent changes an execution traject…
Multi-modal integrated sensing and communication (ISAC) enables environmental perception a…
Corporate and investment banking teams use presentations to support credit decisions and a…
Single-step retrosynthesis models are commonly evaluated by their ability to recover recor…
When designing machine learning models, desirable properties are often in tension: improvi…
Scientific software presents a demanding test for computer-using agents based on visual la…
Grammar-constrained generation guarantees syntactic validity, but can substantially degrad…
Exploratory reinforcement learning (RL) on an operating bus fleet is impractical,while pol…
Large language models frequently fail to balance staying truthful with being supportive. T…
A small, fast decision model, or system one model, for agentic workflows. Pick between opt…
an open-source, from-first-principles reconstruction of the ideas behind TypeSafe AI's Jev…
Blackstone's Jas Khaira will take the Builders Stage at TechCrunch Disrupt 2026 on buildin…
Clay Co-founder and CEO Kareem Amin joins the AI Stage to discuss the rise of GTM engineer…
Local AI is becoming more useful by the token. As AI agents move from experiments into eve…
让每一次请求选对模型，让每一次反馈都成为下一次更优、更省的选择
44年前被亲自列入问题清单
GPT-6 Astra Ultrafast, running on NVIDIA Blackwell GPUs, is available now in the OpenAI AP…
President Trump reportedly asked for Grok's opinion before invading Venezuela and capturin…
OpenAI is rolling out new shopping features for ChatGPT that let users virtually try on cl…
OpenAI has parted ways with three safety researchers after an internal investigation found…
Opus 5.5’s biggest tell is the word “dependable,” which pops up 23 times more often than i…
Advanced AI may matter most for the routine work behind breakthrough ideas. Explore why ex…
Amazon Web Services' Strand Labs has released the latest Jevalike decision model, Strands …
Shopify’s new Canvas site builder lets merchants create and customize their online stores …
Albertsons Cos. is using ChatGPT Enterprise and the OpenAI API to help teams work faster a…
Brian Chesky on making Airbnb agent-friendly, the state of consumer AI, and why the world …
用ImageNet训练encoder
价格只有Astra一半
As it opens a new location, the social club prepares grant applications in 2 hours instead…
YouTube · Google for Developers
YouTube · Google for Developers
YouTube · OpenAI