AndroidDaily supplies 350 verifiable tasks on 94 closed-source Android apps evaluated by GRADE (87.37% human agreement), with the strongest model achieving 62% success.
Agentprm: Process reward models for llm agents via step-wise promise and progress
8 Pith papers cite this work. Polarity classification is still indexing.
8
Pith papers citing it
citation-role summary
background 1
method 1
citation-polarity summary
years
2026 8representative citing papers
DataPRM is an environment-aware generative process reward model that improves LLM data analysis agents by 7-11% on benchmarks via active verification and reflection-aware ternary rewards.
GUIDE decomposes GUI agent evaluation into trajectory segmentation, subtask diagnosis, and overall summary to deliver higher accuracy and structured error reports than holistic baselines.
SCPO recovers step-level credit from successful siblings within rollout groups to reduce semantic inconsistency in group-based RL for LLM agents, matching or exceeding baselines on ALFWorld and WebShop.
A two-stage probe (hidden-state estimate plus attention-based correction) yields per-step GRPO rewards that survive prefix contamination and beat external-judge and tree-search rewards in the reported benchmarks.
Longer action horizons bottleneck LLM agent training through instability, but training with reduced horizons stabilizes learning and enables better generalization to longer horizons.
LLM post-training is unified as off-policy or on-policy interventions that expand support for useful behaviors, reshape policies within reachable states, or consolidate behavior across training stages.
citing papers explorer
-
AndroidDaily: A Verifiable Benchmark for Mobile GUI Agents on Real-World Closed-Source Applications
AndroidDaily supplies 350 verifiable tasks on 94 closed-source Android apps evaluated by GRADE (87.37% human agreement), with the strongest model achieving 62% success.
-
Rewarding the Scientific Process: Process-Level Reward Modeling for Agentic Data Analysis
DataPRM is an environment-aware generative process reward model that improves LLM data analysis agents by 7-11% on benchmarks via active verification and reflection-aware ternary rewards.
-
GUIDE: Interpretable GUI Agent Evaluation via Hierarchical Diagnosis
GUIDE decomposes GUI agent evaluation into trajectory segmentation, subtask diagnosis, and overall summary to deliver higher accuracy and structured error reports than holistic baselines.
-
Semantic Consistency Policy Optimization for Reinforcement Learning of LLM Agents
SCPO recovers step-level credit from successful siblings within rollout groups to reduce semantic inconsistency in group-based RL for LLM agents, matching or exceeding baselines on ALFWorld and WebShop.
-
PAIR: Prefix-Aware Internal Reward Model for Multi-Turn Agent Optimization
A two-stage probe (hidden-state estimate plus attention-based correction) yields per-step GRPO rewards that survive prefix contamination and beat external-judge and tree-search rewards in the reported benchmarks.
-
On Training Large Language Models for Long-Horizon Tasks: An Empirical Study of Horizon Length
Longer action horizons bottleneck LLM agent training through instability, but training with reduced horizons stabilizes learning and enables better generalization to longer horizons.
-
Large Language Model Post-Training: A Unified View of Off-Policy and On-Policy Learning
LLM post-training is unified as off-policy or on-policy interventions that expand support for useful behaviors, reshape policies within reachable states, or consolidate behavior across training stages.
- From Reasoning to Agentic: Credit Assignment in Reinforcement Learning for Large Language Models