information-flow

Tag

Cards List
#information-flow

Faithfulness as Information Flow: Evaluating and Training Faithful Chain-of-Thought Reasoning

arXiv cs.LG · 2026-05-26 Cached

This paper proposes a framework to evaluate and improve faithfulness of chain-of-thought reasoning by controlling information flow, using entropy-based, KL-divergence, and gradient-based diagnostics, and introduces training interventions (attention masking, gradient masking, adversarial perturbations) that make reasoning more transparent and reduce shortcut reliance.

0 favorites 0 likes
#information-flow

@WEB3_furture: How to efficiently obtain the most valuable information streams from both Chinese and English sources? These two free and open-source tools are widely recommended by the community. They can automatically capture hot topics across the entire Chinese and English internet, use AI to filter high-value content, generate translated and in-depth analysis briefs, and push them with one click. 1. TrendRadar (58k+ stars) - Chinese region: automatically captures Zhihu, Douyin, ...

X AI KOLs Timeline · 2026-05-24 Cached

Recommend two free and open-source tools: TrendRadar (Chinese region, 58k+ stars) and Horizon (English region, 4.6k+ stars). They can automatically capture hot topics across the entire internet, use AI to filter high-value content, generate translated and analytical briefs, and push them with one click to multiple platforms.

0 favorites 0 likes
#information-flow

Rethinking Cross-Layer Information Routing in Diffusion Transformers

Hugging Face Daily Papers · 2026-05-20 Cached

This paper proposes Diffusion-Adaptive Routing (DAR), a learnable, timestep-adaptive residual replacement that improves cross-layer information flow in Diffusion Transformers, leading to significant training acceleration and quality improvements.

0 favorites 0 likes
#information-flow

Follow the Flow: On Information Flow Across Textual Tokens in Text-to-Image Models

arXiv cs.CL · 2026-04-20 Cached

This paper investigates how semantic information is distributed across textual tokens in text-to-image models, finding that information concentration and cross-item interactions significantly affect image generation alignment. The authors use patching techniques to demonstrate that simple encoding-stage interventions can improve alignment quality.

0 favorites 0 likes
← Back to home

Submit Feedback