arXiv AI
Counterfactual Regret for Reflective Agents with Lean-Verified Guarantees
Read the original on arXiv AI →The Flow has not summarised this story yet — read it at arXiv AI.
Related stories
arXiv AI
Oct 1
Trust the Critic More
The paper introduces Actor‑Critic with Action Chunking (AC2), a method that assigns credit to short action chunks instead of entire trajectories, enabling policy updates without waiting for terminal rewards. AC2 employs local readiness, reference solutions, and 10k‑token chunks to make critic‑based credit assignment reliable. Experiments on Qwen3‑4B with FineProofs‑RL show AC2 surpasses GRPO’s peak validation score while using 2.5× fewer decoding FLOPs and fewer training steps.
arXiv Computation and Language
3d ago
Beyond Self-Reflection: An Execute-Distill-Verify Framework for Agentic Experience Learning
The paper introduces EDV, an Execute‑Distill‑Verify framework that constructs offline experience for large language model agents without updating model parameters. In EDV, multiple heterogeneous agents execute tasks, a distiller extracts candidate lessons from their trajectories and feedback, and executors review these lessons, keeping only those unanimously approved. Experiments on τ²‑bench, Mind2Web, and MMTB demonstrate improved performance on revisited tasks and within‑domain transfer, while a Retail subset evaluation shows EDV cuts the Self‑Confirmation Trap Rate from 11.0% to 5.0%.
arXiv AI
5d ago
Verify--Backtrack Loop: Structured Precondition Recovery for Long-Horizon Language Agents
arXiv:2610.41564v1 Announce Type: new Abstract: Language agents increasingly tackle long-horizon tasks in interactive environments, where actions often depend on latent preconditions. The core diffic...
arXiv AI
Sep 7
Forgetting Without Restarting: Execution-State Unlearning for Stateful LLM Agents
arXiv:2609. 04875v1 Announce Type: cross Abstract: Long-running LLM agents are stateful: beyond the transcript they accrete compressed summaries, plaintext memory, pending tool plans, and, under every serving API, a KV cache.
arXiv AI
Sep 30
Counterfactual Rollout Replay: Forkable Environments as Free Process Rewards for Software Engineering Agents
arXiv:2609.33875v2 Announce Type: replace-cross Abstract: Outcome-only reinforcement learning gives software engineering (SWE) agents a terminal success signal but little direct guidance about interm...
arXiv AI
Aug 20
Training-Free Inference-Time Self-Reflection and Cost-Bounded Early Stopping for Large Language Models
The paper introduces EvoResearcher, a training‑free, inference‑time protocol that enables a frozen large language model to perform cost‑bounded self‑reflection and early stopping. By iterating through generate → self‑critique → revise steps until a maximum depth or a CONFIRMED sentinel is reached, the model can self‑verify its answers within a strict compute budget. The protocol incorporates four self‑reflective meta‑reward components—correctness, efficiency, reflection depth, and tool‑call diversity—implemented as prompt‑level mechanisms, and is validated on Big‑Bench Hard, GSM8K, and MATH benchmarks, achieving comparable accuracy while terminating 82‑88% of items early with only about 2.1 generations per question.