@sheriyuo: latent reasoning 一直是个很不错的研究方向，从 soft prompting 就能看到苗头 如果真的能带来更高的 bound，那么过错的不是这个 blackbox，而是做不到 latent efficiency 的 res…
Summary
讨论latent reasoning作为研究方向的价值，并质疑OpenAI o1模型宣称的thinking过程是否真实或只是对外宣传，认为模型可能尚未达到上限。
View Cached Full Text
Cached at: 07/03/26, 12:28 AM
latent reasoning 一直是个很不错的研究方向，从 soft prompting 就能看到苗头
如果真的能带来更高的 bound，那么过错的不是这个 blackbox，而是做不到 latent efficiency 的 researcher
Panda (@Jiaxi_Cui): 当前模型已经到上限了吗？可能还没有
让模型自己 thinking 这个事从 o1 开始我就不太信，考虑到现实情况，更多应该是当时的 OpenAI 需要向外界 claim 自己在模型上还可以走的很远，所以要做这个很容易让外人理解的 thinking 的输出结构
但抛开这样的原因，为什么模型要用 nature language 的形式来
Similar Articles
@Phoenixyin13: Finished reading a long post today by OpenAI researcher Noam Brown — a reality severely underestimated by the industry. The true ceiling of LLM capabilities is far higher than what any current benchmark shows. The reason: too little test-time compute. And as models...
Highlights OpenAI researcher Noam Brown's argument: the true ceiling of LLM capabilities is far higher than current benchmarks show, due to insufficient test-time compute, and stronger models benefit more from additional computation. This poses a serious challenge for AI safety evaluation, as many dangerous capabilities may only emerge under long time and high compute budgets.
@ZeroZ_JQ: https://x.com/ZeroZ_JQ/status/2079504278922891568
This article provides an in-depth explanation of the principle of reasoning effort in large language models: by adjusting the reasoning effort level (low/medium/high) on the same model, more intermediate reasoning trajectories (serial autoregressive computation) are allowed before outputting the final answer, thereby making trade-offs among answer quality, response speed, and computational cost. The key is that the model has learned to organize longer generation processes into effective computation, rather than simply increasing parameters or network layers.
Economics and reasoning with OpenAI o1
OpenAI released the o1 model series, designed with extended reasoning capabilities to tackle complex problems in science, coding, and math by spending more time thinking before responding.
Reasoning models struggle to control their chains of thought, and that’s good
OpenAI researchers study whether reasoning models can deliberately obscure their chain-of-thought to evade monitoring, finding that current models struggle to control their reasoning even when aware of monitoring. They introduce CoT-Control, an open-source evaluation suite with over 13,000 tasks to measure chain-of-thought controllability in reasoning models.
@burny_tech: A Survey on Latent Reasoning "Large Language Models (LLMs) have demonstrated impressive reasoning capabilities, especia…
This survey provides a comprehensive overview of latent reasoning in LLMs, exploring methods that perform multi-step inference in continuous hidden states without explicit token-level supervision.