Chain-of-Thought (CoT)
PulseAugur coverage of Chain-of-Thought (CoT) — every cluster mentioning Chain-of-Thought (CoT) across labs, papers, and developer communities, ranked by signal.
4 day(s) with sentiment data
-
CoT fine-tuning degrades LLM long-context recall; QK-Restore fixes it
Researchers have identified a significant issue where Chain-of-Thought (CoT) fine-tuning, intended to boost reasoning, inadvertently harms the long-context recall capabilities of hybrid linear-attention models. This deg…
-
LLMs improve reasoning with new Verification-First prompting strategy
Researchers have developed a new prompting strategy called Verification-First (VF) to improve Large Language Model reasoning without significant training costs or extensive sampling. This method prompts LLMs to verify a…
-
LLM reasoning effectiveness predicted by entropy dynamics
Researchers have developed a new framework called EDRM that uses early-stage entropy dynamics to determine when Large Language Models (LLMs) should engage in explicit reasoning. They observed that tasks benefiting from …
-
Small LLMs use positional copying shortcut for arithmetic, bypassing CoT logic
A new research paper reveals a significant shortcut in how small language models perform arithmetic tasks using chain-of-thought (CoT) prompting. Instead of relying on logical sequencing, these models tend to copy the n…