Research map / Vision-language and multimodal / Robot action and navigation

Chain-of-Correction: Progress-Aware Policy Steering via Anchor-Grounded Predictive Reasoning

NeurIPS 2026 · Vision-language and multimodal › Robot action and navigation · approach: Memory and subgoal-based manipulation

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes Chain-of-Correction, an inference-time policy steering method that evaluates and refines imitation-learning action proposals via anchor-grounded predictive reasoning for robotic manipulation.

Problem

Imitation learning policies for robotic manipulation suffer compounding errors; existing inference-time policy steering anticipates consequences in unstructured pixel or latent spaces, producing physically inconsistent predictions and unreliable error detection.

Approach

Replaces unstructured pixel or latent-space outcome prediction with anchor-grounded predictive reasoning about task progress to evaluate and refine action proposals before execution (abstract only; details not available).

Experimental setup

Builds on: policy steering

Closest papers by meaning

• Chain-of-Action: Trajectory Autoregressive Modeling for Robotic Manipulation (NeurIPS 2025 · similarity 0.90)
• Phoenix: A Motion-based Self-Reflection Framework for Fine-grained Robotic Action Correction (CVPR 2025 · similarity 0.89)
• FlowPolicy: Enabling Fast and Robust 3D Flow-Based Policy via Consistency Flow Matching for Robot Manipulation (AAAI 2025 · similarity 0.89)
• PACE: Phase-Aware Chunk Execution for Robot Policies with Action Chunking (NeurIPS 2026 · similarity 0.89)
• Implicit Drifting Policy: One-Step Action Generation via Conditional Expert Geometry (NeurIPS 2026 · similarity 0.89)
• LaST-R1: Reinforcing Robotic Manipulation via Adaptive Physical Latent Reasoning (NeurIPS 2026 · similarity 0.88)