arxiv arXiv cs.AI · 6h ago · research

Atria Dawn Preview: foundation agentic language model for scientific research

from English

Atria Dawn Preview is a foundation agentic language model designed for scientific research and engineering workflows, trained via a Verifiable Experience Pipeline that connects tool-mediated interactions to executable environments. Across 16 benchmarks spanning real-world research, engineering, and digital work, the model is competitive with frontier agents and achieves the highest reported score on five of them.

The training pipeline connects tool-mediated interactions to executable environments and externally verified outcomes.
Analysis of 769 task records from 56 participants shows that about one-third of completed AI-assisted tasks were rated as infeasible without AI.
Agents frequently propose methods and implement revisions, while humans retain most final decisions and guide exploration through judgment and feedback.

The authors consider this significant as it indicates a shift from task-level execution to project-level partnership, where human effort concentrates on what is worth pursuing and how evidence should guide research.

Importance 2/3 arXiv cs.AI DeepSeek AI agents Research paper

Related articles

arxiv arXiv cs.CL · 20d ago · 25 views

JIT-Agent enables on-the-fly harness evolution to boost agentic LLM performance

The authors present JIT-Agent, a harness intelligence model trained to synthesize task-adaptive agent harnesses for arbitrary off-the-shelf agentic LLMs. This approach formalizes the agent harness as a composable artifact governed by a fixed four-module protocol, allowing the system to customize and repair harnesses on the fly.

arxiv arXiv cs.CL · 29d ago · 16 views

Mint-Agent introduces finance-native agentic foundation models

Researchers present Mint-Agent, a family of finance-native agentic models designed to combine reliable execution with long-horizon research capabilities. The system is built on three pillars: a data engine for specialized tasks, the MintHarness environment for auditable evidence trails, and a training recipe combining SFT, OPD, and RLVR.

arxiv arXiv cs.LG · 32d ago · 36 views

Faraday 27B outperforms Claude Opus 4.8 and GPT-5.5 on paper replication tasks

Researchers have developed Replica, a scalable task space for paper replication, to address the need for reliable scientific knowledge. They post-trained Faraday, a 27B-parameter "AI Scientist" agent that uses coding agents as tools, to leverage this environment.

arxiv arXiv cs.AI · 32d ago · 29 views

Faraday 27B surpasses Claude Opus 4.8 and GPT-5.5 on paper replication tasks

Researchers have developed Replica, a scalable task space for paper replication, and introduced an auto-generated rubric-based judge to provide low-noise reward signals that align with human assessment.

arxiv arXiv cs.CL · 73d ago · 22 views

AutoMem automates learning of memory as a cognitive skill for LLMs

Researchers introduce AutoMem, a framework that treats memory management in large language models as a trainable skill rather than a static component. The system promotes file-system operations to first-class memory actions, allowing the model to autonomously decide how to encode, retrieve, and organize knowledge.