Search: “reasoning”

arXiv:2603.18256v2 Announce Type: replace-cross Abstract: Recent reasoning-based large language models have shown strong performance on tasks with verifiable outcomes, but their use in de novo molecul

arXiv:2604.17574v1 Announce Type: new Abstract: Distractor generation (DG) remains a labor-intensive task that still significantly depends on domain experts. The task focuses on generating plausible y

arXiv:2604.16278v1 Announce Type: new Abstract: Although most of the automated theorem-proving approaches depend on formal proof systems, informal theorem proving can align better with large language

arXiv:2510.26109v4 Announce Type: replace Abstract: Reinforcement learning with verifiable rewards (RLVR) has significantly boosted the reasoning capability of language models (LMs). However, existing

arXiv:2604.13371v1 Announce Type: new Abstract: Large Language Models (LLMs) are increasingly described as possessing strong reasoning capabilities, supported by high performance on mathematical, logi

arXiv:2505.04842v2 Announce Type: replace-cross Abstract: Prevalent reinforcement learning~(RL) methods for fine-tuning LLM reasoners, such as GRPO or Leave-one-out PPO, abandon the learned value func

A lot of real-world documents embed extremely dense charts that require humans to not only 'perceive' the chart values, but also reason over them and reconcile with other complex elements (e.g. tables

arXiv:2610.21392v1 Announce Type: new Abstract: Reinforcement learning from internal feedback can improve language models without ground-truth answers or external verifiers. However, confidence and co

arXiv:2610.00825v1 Announce Type: new Abstract: Dubbing quality control requires a reference-free judge that can determine whether a candidate text line matches a speaker's visible articulation in bot

arXiv:2609.38340v1 Announce Type: new Abstract: When a materials LLM answers a question about crystal structure, does it reason from the structure or copy an answer already printed in its input? Accur

arXiv:2609.34857v1 Announce Type: new Abstract: Reinforcement learning with verifiable rewards (RLVR) trains reasoning models to produce correct answers, but does not ensure that their stated confiden

arXiv:2609.33399v1 Announce Type: new Abstract: In realistic education, a solution is often expressed not only in words but in a drawing--a circuit, a geometric construction, a function plot--and a te

arXiv:2609.18958v1 Announce Type: cross Abstract: The natural way to review a long recording or document with a multimodal model is to hand it the raw source and ask for a review in one call. We show

arXiv:2609.18677v1 Announce Type: new Abstract: Speech language models enable richer spoken interactions between humans and machines than cascaded systems, allowing access to paralinguistic informatio

arXiv:2609.16268v1 Announce Type: new Abstract: Large language model (LLM) agents increasingly interleave natural language reasoning with external tools such as web search and code execution. These to

arXiv:2609.12606v2 Announce Type: replace Abstract: While multimodal reasoning has advanced rapidly, solving complex geometry problems critically hinges on active visual assistance, such as constructi

During an outage, every second matters. That is one of the reasons OpenAI is using @cerebras internally. @seanlie describes how fast inference is being used for incident response and critical research

Hi everybody, we post-trained Qwen 3.8 27B to be more efficient by figuring out which tokens were linked to overthinking and penalizing them without 'attacking' the reasoning length directly then fixe

arXiv:2606.21704v2 Announce Type: replace Abstract: Chain-of-Thought (CoT) distillation transfers multi-step reasoning from large reasoning models to smaller students, but verbose teacher traces infla

arXiv:2608.31066v1 Announce Type: new Abstract: Chain-of-thought (CoT) reasoning improves multi-step problem solving, but long reasoning traces inflate inference cost. Token-level CoT compression redu

I'm in San Francisco this week. So is the Deputy Prime Minister, but we’re here for very different reasons. He's meeting with Anthropic and OpenAI, working on Australia’s data centre pipeline. I'm her

Introducing Hy4 Preview New open weight text input (no vision) LLM from Chinese company Tencent today: 770B total parameters, 49B active parameters, 1M token context window, 1.56TB on Hugging Face. Th

arXiv:2608.26596v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) are increasingly capable scientific assistants, yet they remain far from fully autonomous research. This transi

Intermediate tokens, called 'thinking' or 'reasoning' actually are nothing like it. Humans do step-by-step reasoning leading to the conclusion. LLMs use intermediate traces to augment their prompt. Th

arXiv:2608.16443v1 Announce Type: new Abstract: Neurosymbolic (NeSy) Artificial Intelligence aims to integrate Deep Learning (DL) architectures with symbolic reasoning. While initial NeSy approaches h

'Again?' you might ask. I(*) took the original Qwen 27B 3.8 chat template and compared it to the improved template that was posted a day before. The original has issues. The improved one fixed some wh

arXiv:2608.05687v1 Announce Type: cross Abstract: Masked diffusion language models (dLLMs) can commit tokens in any order -- a freedom marketed as their core advantage over autoregressive decoding. We

arXiv:2608.03263v1 Announce Type: cross Abstract: We test whether the 'compositional ignition' reported in latent-reasoning models is real computation, an instrument artifact, or inherited from verbal

arXiv:2607.22947v1 Announce Type: new Abstract: A formal model enables verifying reachability, localizing an outage, or anticipating the blast radius of a change. Yet, virtually no production network

Was very cool to hear about the reasons people love Sol. We're doing the promotion again, except this time for ChatGPT Work: Tweet what you love about ChatGPT Work, claim 100 in free credits, get more

arXiv:2607.13899v1 Announce Type: new Abstract: We propose the AIMO Interpretability Challenge, a competition on distinguishing robust from spurious reasoning in frontier mathematical language models

arXiv:2606.29296v1 Announce Type: new Abstract: Group Relative Policy Optimization (GRPO) is a default recipe for process-supervised reinforcement learning of LLM reasoners, and dense process supervis

This, once again, shows why the theory of change of 'we just wait for a wArNiNg ShOt and then everyone will suddenly be reasonable!' is flawed. We need to do a lot of work to make the situation legibl

arXiv:2511.10234v3 Announce Type: replace-cross Abstract: While promising, graph reasoners based on Large Language Models (LLMs) lack built-in invariance to symmetries in graph representations. Operat

arXiv:2601.15158v4 Announce Type: replace-cross Abstract: Transformers trained via Reinforcement Learning (RL) with outcome-based supervision can spontaneously develop the ability to generate intermed

arXiv:2602.08868v2 Announce Type: replace-cross Abstract: Time-series anomaly detection (TSAD) with multimodal large language models (MLLMs) is an emerging area, yet a persistent challenge remains: ML

arXiv:2606.01215v1 Announce Type: cross Abstract: Current 3D spatial reasoning methods face a fundamental trade-off: neuro-symbolic 3D (NS3D) concept learners achieve interpretable reasoning through c

arXiv:2606.00476v1 Announce Type: new Abstract: Do LLM agents act on the reasoning they state? This question of process fidelity is central to using LLMs in social simulation, yet it is hard to measur

arXiv:2602.02518v2 Announce Type: replace-cross Abstract: Large language models (LLMs) increasingly rely on external knowledge to improve factuality, yet many real-world knowledge sources are organize

arXiv:2605.24004v1 Announce Type: new Abstract: Large language models (LLMs) are promising for autonomous driving, but semantics-only decision policies can yield physically unsafe behavior in dynamic

arXiv:2605.17088v1 Announce Type: new Abstract: Recent advances in large language models (LLMs) have shown that Chain-of-Thought (CoT) reasoning can substantially improve performance on complex reason

arXiv:2602.02834v3 Announce Type: replace-cross Abstract: What structural inductive bias helps transformers reason over knowledge graphs? Through controlled ablations of a minimal transformer modifica

This is confused, but popular. Popular because it tells a bunch of people what they want to hear. Confused for a couple reasons: first, Mythos probably isn’t a pure LLM. (Claude Code isn’t, and it pro

arXiv:2605.04396v1 Announce Type: new Abstract: Recent work has shown that Transformers' compositional generalization is governed by complexity control, initialization scale and weight decay, which st

arXiv:2512.10941v2 Announce Type: replace-cross Abstract: Reasoning goes beyond language; the real world requires reasoning about space, time, affordances, and much more that words alone cannot convey

arXiv:2604.26573v1 Announce Type: new Abstract: Improving large language model (LLM) reasoning requires supervision that is both aligned with the model's own test-time states and informative at the to