Agents

Scaling Scientific Discovery Environments for Turn-Level Agentic RL

arXiv:2607.28990v1 Announce Type: new Abstract: Large language model agents have shown promising capabilities in data-driven scientific discovery tasks, where an agent interacts with an execution envi

DGX agentpaper
agentsarxiv-cs-ai

arXiv:2607.28990v1 Announce Type: new Abstract: Large language model agents have shown promising capabilities in data-driven scientific discovery tasks, where an agent interacts with an execution environment and produces a statistical claim. Long-horizon scientific analysis remains constrained by the lack of process supervised environments over real-world scientific data. This paper introduces SciDisco, a scalable framework for training Scientific Discovery agents in process-verifiable environments. SciTheque compiles hypotheses, datasets, hidden evidence graphs, and verifiers into task environments where analytical progress can be checked during interaction. DAG-grounded trajectory synthesis uses these environments to construct verifier-filtered multi-turn demonstrations. DiscoPO then uses the environment as the source of training signal, assigning turn-level credit to actions that produce verifiable analytical evidence. Experiments show that SciDisco-14B reaches state-of-the-art on hypothesis-driven scientific data analysis benchmarks.

Related

Source: arXiv cs.AI | 2026-08-03

Loading related sources…