7. AdaptThink This paper introduces AdaptThink, an RL framework designed to help reasoning models decide when to use detailed chain-of-thought reasoning (“Thinking”) versus directly producing an answer (“NoThinking”), based on task difficulty.
PROMPT ENGINEERING
-
Chain-of-Thought Prompting Trade-offs in Reasoning Models
By
–
4. The Pitfalls of Reasoning Explores an unexpected flaw in reasoning-augmented large language models (RLLMs): while chain-of-thought (CoT) prompting often boosts performance on complex reasoning tasks, it can degrade instruction-following accuracy.
-
Novel LLM Training Approach for Thoughtful Evaluation Reasoning
By
–
3. J1 Introduces a novel training approach for LLMs to act as evaluators (LLM-as-a-Judge) by explicitly incentivizing thoughtful reasoning during judgment.
-
Hands-On Large Language Models: Complete Code and Notebooks Guide
By
–
Want to get hands-on with Large Language Models? This repo has all the code from the `Hands-On Large Language Models` book , covering everything from prompt engineering to fine-tuning! 12 chapters of practical notebooks: ↳ Chapt. 1: Introduction to Language Models
↳ -
Exploring Creative AI: Iteration Without Language Barriers
By
–
I want to explore, create and iterate without always having to find the words to do it.
-
Claude 4 Opus Solves Problem After 4 Hints, o3 Shows Improvement
By
–
fyi for anyone interested later, e.g. the new Claude 4 Opus gets there after 4 hints https://
claude.ai/share/33072dd0
-a760-4b0f-8b73-6c0bb4a883ad
…
Other LLMs do similar except – o3 didn't get it yesterday but when I tried this morning it did and now I can't tell if that's just due to the new conversation memory -
AI Generation Succeeds Without Expected Content Denial
By
–
was taking screenshots during this generation thinking it would get denied but… it worked fine
-
System thinking mode reliability improvements and user warnings
By
–
1. i am reporting what others have shared. 2. one would have to do more experiments to see whether “thinking mode’ reliably fixes it. 3. the system should warn you and tell you to use thinking mode if that does reliably fix it.
-
Agents Autonomes Optimisent Engagement Via Boucles Feedback Récursives
By
–
Agents are quite good honestly. I played around with prompts to try to post engagement bait, study my past tweets that got high engagement, and optimize for replies and likes. I think letting agents run reliably with full autonomy and recursive feedback loops is around the