@yoheinakajima: i don’t get it fully but this looks cool
Summary
Yohei Nakajima shares a breakthrough from Bad Theory Labs where they improved LLM reasoning to overcome linear thinking, making it more adaptable.
View Cached Full Text
Cached at: 09/26/26, 09:01 PM
i don’t get it fully but this looks cool
Bad theory labs (@Badtheorylabs): We achieved a breakthrough at BTL.
AI models are very linear. They think in one direction at a time, and when they’re wrong they scramble and start again. For something meant to replace humans, that’s way too human.
We changed that. We made LLM reasoning and execution work more
Similar Articles
@mdancho84: BREAKING: MIT researchers discover how to enable LLMs to do real logical reasoning. This is what you need to know: (thr…
MIT researchers have discovered a method to enable large language models to perform real logical reasoning, as announced in a Twitter thread.
@rao2z: "When an LLM outputs a step-by-step plan, it creates a powerful illusion that you are watching a machine reason its way…
A position paper by Subbarao Kambhampati and researchers at Arizona State University argues that chain-of-thought reasoning in LLMs creates an illusion of reasoning, and the industry needs to move beyond costly token generation to alternative reasoning mechanisms.
A*-Thought-V2: Efficient Latent Reasoning via Geometric Dynamics of LLM
A*-Thought-V2 models chain-of-thought reasoning as hidden-state trajectories, using geometric dynamics to compress non-essential steps into latent tokens, improving accuracy and efficiency in LLM reasoning.
Why LLMs Hallucinate on Structured Knowledge: A Mechanistic Analysis of Reasoning over Linearized Representations
This paper presents a mechanistic analysis of why LLMs hallucinate when reasoning over linearized structured knowledge, finding that hallucinations stem from systematic internal dynamics such as attention on shortcut cues and failures in semantic grounding in feed-forward layers, rather than random noise.
@mdeng34: Frontier LLMs are converging on efficient, adaptive reasoning. Opus 4.7 lets the model decide how deeply to reason. GPT…
New research introduces SR²AM, a configurator that self-regulates when to use simulative reasoning, improving efficiency and performance in LLMs.