ICML 2026 | 多模态思维链真的靠谱吗？川大新框架拒绝噪声思维干扰
Ranking
Overall
74
Content
80
Popularity
59
Observed public metrics from 1 member.
Merged summary
TL;DR - An ICML 2026 paper from Sichuan University, "Reliable Thinking with Images" (RTWI), tackles "noisy thoughts" in multimodal chain-of-thought reasoning, where wrong visual cues or faulty reasoning cascade into wrong answers. It offers a plug-and-play test-time scaling framework that raises accuracy while cutting inference cost.
- Frames the Thinking-with-Images (TWI) pipeline as two stages — cue mining (tool-driven visual clue extraction) and answer reasoning — and shows via error analysis that incorrect answers usually trace back to errors in one of these stages.
- Instead of modeling continuous visual uncertainty directly, RTWI uses a text-centric reliability measure based on token entropy per stage, since the tool-calling text instruction is the precondition for obtaining correct clues.
- Two empirical observations drive the method: reliability correlation (higher stage reliability ↔ higher answer accuracy) and reliability jump (correct visual clues yield larger reliability gains from cue mining to answer reasoning).
- The framework combines dual-stage filtering (percentile thresholds discard unreliable paths) with reliability-weighted voting replacing majority voting; tested on Qwen3-VL and DeepEyes across high-resolution, TWI-specific, multimodal math, and open-ended VQA benchmarks, reporting accuracy gains, competitive token-saving rates via online early stopping, and consistent benefits across model scales.
Sources (1)
ICML 2026 | 多模态思维链真的靠谱吗？川大新框架拒绝噪声思维干扰
Public signals
Semantic Scholar citations 4 · Semantic Scholar influential citations 0
TL;DR - An ICML 2026 paper from Sichuan University, "Reliable Thinking with Images" (RTWI), tackles "noisy thoughts" in multimodal chain-of-thought reasoning, where wrong visual cues or faulty reasoning cascade into wrong answers. It offers a plug-and-play test-time scaling framework that raises accuracy while cutting inference cost.
- Frames the Thinking-with-Images (TWI) pipeline as two stages — cue mining (tool-driven visual clue extraction) and answer reasoning — and shows via error analysis that incorrect answers usually trace back to errors in one of these stages.
- Instead of modeling continuous visual uncertainty directly, RTWI uses a text-centric reliability measure based on token entropy per stage, since the tool-calling text instruction is the precondition for obtaining correct clues.
- Two empirical observations drive the method: reliability correlation (higher stage reliability ↔ higher answer accuracy) and reliability jump (correct visual clues yield larger reliability gains from cue mining to answer reasoning).
- The framework combines dual-stage filtering (percentile thresholds discard unreliable paths) with reliability-weighted voting replacing majority voting; tested on Qwen3-VL and DeepEyes across high-resolution, TWI-specific, multimodal math, and open-ended VQA benchmarks, reporting accuracy gains, competitive token-saving rates via online early stopping, and consistent benefits across model scales.