Official codebase for "Can 1B LLM Surpass 405B LLM? Rethinking Compute-Optimal Test-Time Scaling".
-
Updated
Feb 19, 2025 - Python
#
process-reward-model
Here are 20 public repositories matching this topic...
[ICML 2025] Reward-guided Speculative Decoding (RSD) for efficiency and effectiveness.
efficiency reasoning decoding-algorithm large-language-models speculative-decoding process-reward-model
-
Updated
May 2, 2025 - Python
This repository contains the official implementation of Error Typing for Smarter Rewards: Improving Process Reward Models with Error-Aware Hierarchical Supervision.
-
Updated
May 27, 2025 - Jupyter Notebook
[KDD 2025] Rewarding Graph Reasoning Process makes LLMs more Generalized Reasoners
-
Updated
May 30, 2025 - Python
[NeurIPS 2025 Spotlight] LLM post-training suite — featuring ReasonFlux, ReasonFlux-PRM, and ReasonFlux-Coder.
reinforcement-learning code-generation post-training chain-of-thought llm-rlhf gemini-pro sft-data process-reward-model deepseek-r1 o3-mini clawdbot-skill
-
Updated
Sep 27, 2025 - Python
[AAAI 2026] Official codebase for "GenPRM: Scaling Test-Time Compute of Process Reward Models via Generative Reasoning".
-
Updated
Nov 8, 2025 - Python
Interface for DeepSeek-R1 chain-of-thought model — exposes the internal reasoning trace, supports API and local Ollama backends, with streaming output and temperature control.
deep-learning neural-networks reasoning large-language-models llm rlhf reinforcement-learning-from-human-feedback chain-of-thought-reasoning deepseek process-reward-model
-
Updated
Mar 15, 2026
Live proof of arXiv:2603.17815 — O(N) confirmed R²=0.952, 1,984 API calls
-
Updated
Mar 19, 2026 - Python
This repository includes code and materials for the paper "Efficient PRM Training Data Synthesis via Formal Verification" (ACL 2026 Findings).
-
Updated
Apr 7, 2026 - Python
Step-level preference data construction toolkit for code-agent process reward models
-
Updated
May 18, 2026 - Python
Fixing GRPO training collapse in long-horizon multi-tool agents. A lightweight PRM-Lite + LATA joint approach achieves +37% over vanilla GRPO on τ-bench airline (50-task, multi-turn).
reinforcement-learning long-horizon qwen agentic-ai tool-calling process-reward-model grpo tau-bench multi-turn-agents
-
Updated
Jun 27, 2026 - Python
[ACL 2026 SAC Highlight Award] Can We Predict Before Executing Machine Learning Agents?
agent machine-learning natural-language-processing rollout mlagent large-language-models process-reward-model
-
Updated
Jul 7, 2026 - Python
DPO-trained decision layer for coding agents — raises pass@1 from 0.419 → 0.639 on frozen 14B generator (p<0.000001)
code-generation fine-tuning dpo clarification humaneval llm qlora qwen coding-agents process-reward-model
-
Updated
Jul 28, 2026 - Python
Multi-Agent System Process Reward Model (MASPRM): a lightweight process reward model guiding multi-agent systems at search time.
-
Updated
Aug 22, 2026 - Python
Code for 'Characterizing a Configuration Where Inference-Time PRM-Pruned Fragment Grafting Is Inert: Evidence from Three Reasoning LMs' — a rigorously established inference-time mechanism null (3 LMs, 6 benchmarks, 2 PRMs, N=8–256, pre-registered TOST)
-
Updated
Aug 27, 2026 - Python
A comprehensive collection of process reward models.
-
Updated
Sep 13, 2026
Official code for CulturePRM: A Process Reward Model for Mitigating Cultural Overriding in Cultural Reasoning (EMNLP 2026 Main)
-
Updated
Sep 20, 2026 - Python
Curated papers, taxonomy, benchmarks, and decision guides for credit assignment in reasoning and agentic LLM reinforcement learning.
awesome reinforcement-learning awesome-list multi-agent-systems credit-assignment large-language-models llm process-reward-model rlvr agentic-rl
-
Updated
Sep 23, 2026 - Python
Add this topic to your repo
To associate your repository with the process-reward-model topic, visit your repo's landing page and select "manage topics."