Newest papers

Self-Credit: Credit Assignment for Long-Horizon Agentic RL via Intra-group Rescaling

Fangcong Yin, Ming Yin, Tanvi Bhandarkar, Aashu Singh, Eryk Helenowski, Shlok Kumar Mishra, Saghar Hosseini, Karthik Abinav Sankararaman · 2026 · arXiv

DOI: 10.48550/arXiv.2610.28388 · open-access PDF

Indonesian

Agen LLM biasanya dilatih dengan RLVR, tetapi sifat yang rapuh dari reward RLVR khas membuat alokasi kredit menantang, terutama untuk tugas jangka panjang. Kami mengusulkan Self-Credit, metode agen RL end-to-end yang menciptakan sinyal pengawasan padat dari reward hasil yang rapuh menggunakan alokasi kredit yang dihasilkan sendiri melalui rescaling intra-group. Self-Credit membangun RL berbasis kelompok dengan tiga komponen. Pertama, selama peluncuran, kebijakan model menghasilkan ringkasan dengan kutipan ke putaran sebelumnya yang berisi bukti kritis, yang berfungsi sebagai kredit yang dihasilkan sendiri. Kedua, kami menyempurnakan mo ini

Machine translation (Meta M2M100); the English original follows.

Abstract

LLM agents are commonly trained with RLVR, but the sparse nature of typical RLVR rewards makes credit assignment challenging, especially for long-horizon tasks. We propose Self-Credit, an end-to-end agentic RL method that creates dense supervision signals from sparse outcome rewards using self-generated credit assignment via intra-group rescaling. Self-Credit builds on group-based RL with three components. First, during its rollouts, the policy model generates summaries with citations to earlier turns containing critical evidence, which serve as self-generated credit. Second, we refine this mo

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange