Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach · HackerPapers discussions

Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach — mentioned by dannyw

Mon, 03 Aug 2026 14:32:25 GMT

MoE has nothing to do with multimodality. MoE is a concept proposed in 1991, before the deep learning era (which is before what I call the transformers era). You can think of it like sharing. Contrary to popular belief; 'experts' in MoE LLMs do not specialize. There's no expert trained to be good at maths, or python, or writing, or whatever. It's an inference optimization. As for reasoning in non-text modalities, you might find this paper interesting :) https://huggingface.co/papers/2502.05171

Paper and discussion history. A mention is not an endorsement.

Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach — mentioned by coppsilgold

Tue, 07 Apr 2026 21:56:35 GMT

Likely an improvement on: > We study a novel language model architecture that is capable of scaling test-time computation by implicitly reasoning in latent space. Our model works by iterating a recurrent block, thereby unrolling to arbitrary depth at test-time. This stands in contrast to mainstream reasoning models that scale up compute by producing more tokens. Unlike approaches based on chain-of-thought, our approach does not require any specialized training data, can work with small context windows, and can capture types of reasoning that are not easily represented in words. We scale a…

Paper and discussion history. A mention is not an endorsement.

Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach — mentioned by imranq

Tue, 10 Mar 2026 20:05:48 GMT

Amazing write up and i wish more people showed the process for discovery which is often even more interesting than the result itself Still the result is really interesting being able to stack abstract reasoning and get better performance and the heat maps to show the prob results The academic literature seems to be catching up: - *[SOLAR / DUS (Kim et al., 2023)](https://arxiv.org/abs/2312.15166)* — duplicated transformer layers to build a 10.7B model that outperformed 30B parameter baselines. - *[The Curse of Depth (2025)](https://arxiv.org/abs/2502.05795)* — explains why this works: Pre-LN…

Paper and discussion history. A mention is not an endorsement.

Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach — mentioned by in-silico

Tue, 23 Dec 2025 04:22:22 GMT

> nobody has tried to generalize it for example by combining the recurrence concept with next token prediction Here you go: https://arxiv.org/abs/2502.05171

Paper and discussion history. A mention is not an endorsement.

Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach — mentioned by atrudeau

Wed, 08 Oct 2025 15:33:34 GMT

Very cool to see the effectiveness of recurrence on ARC. For those interested in recurrence, here are other works that leverage a similar approach for other types of problems: Language modeling: Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach https://arxiv.org/pdf/2502.05171 Puzzle solving: A Simple Loss Function for Convergent Algorithm Synthesis using RNNs https://openreview.net/pdf?id=WaAJ883AqiY End-to-end Algorithm Synthesis with Recurrent Networks: Logical Extrapolation Without Overthinking https://arxiv.org/abs/2202.05826 Can You Learn an Algorithm?…

Paper and discussion history. A mention is not an endorsement.

Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach — mentioned by vonneumannstan

Wed, 16 Jul 2025 20:35:32 GMT

Unfortunately I think the competitive pressures to improve model performance makes this kind of monitor-ability is short-lived. It's just not likely that textual reasoning in english ismost optimal. Researchers are already pushing in this direction: https://arxiv.org/abs/2502.05171 "We study a novel language model architecture that is capable of scaling test-time computation by implicitly reasoning in latent space. Our model works by iterating a recurrent block, thereby unrolling to arbitrary depth at test-time. This stands in contrast to mainstream reasoning models that scale up compute by…

Paper and discussion history. A mention is not an endorsement.

Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach — mentioned by korbip

Sat, 15 Feb 2025 03:04:16 GMT

There is a LOT of effort in the research community currently: 1. Improving the Self-Attention in the Transformer as is, keeping the quadratic complexity, which has some theoretical advantage in principle[1]: The most hyped one probably DeepSeek's Multi-Latent Attention[15], which kind of is Attention still - but also somehow different. 2. Linear RNNs: This starts from Linear Attention[2], DeltaNet[3], RKWV[4], Retention[5], Gated Linear Attention[6], Mamba[7], Griffin[8], Based[9], xLSTM[10], TTT[11], Gated DeltaNet[12], Titans[13]. They all have an update like: C_{t} = F_{t} C_{t-1} + i_{t}…

Paper and discussion history. A mention is not an endorsement.

Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach — HN submission

Mon, 10 Feb 2025 19:50:20 GMT

Scaling up test-time compute with latent reasoning: A recurrent depth approach

149 points; 44 discussion replies on HN. Paper and source history. A submission is not an endorsement.

Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach — HN submission

Mon, 10 Feb 2025 08:10:16 GMT

Scaling Up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach

1 point; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.