Chain-of-Thought Reasoning
chain-of-thought-reasoning-7b24a7b8·20 events·first seen 1mo agoAliases: Chain-of-Thought, Chain-of-Thought Reasoning, chain-of-thought visual reasoning
Merged from
Chain-of-Thought
Co-occurring entities
More like this (12)
Guides (1)
Recent events (20)
Probe Trajectories Reveal Reasoning Dynamics in Large Reasoning Models
This paper investigates whether hidden representations of Large Reasoning Models (LRMs) can predict future model behavior by analyzing probe trajectories—the continuous evolution of concept probabilities across Chain-of-Thought reasoning tokens. The authors find that temporal trajectory features (volatility, trend, steady-state) significantly outperform single static probes, with max-pooling achieving up to 95% AUROC across safety and mathematics domains. Two methodological insights are offered: template-based training data matches dynamically generated responses in quality, and pooling strategy is critical to probe performance. The work positions probe trajectories as a complementary safety monitoring framework for LRMs where CoT faithfulness cannot be assumed.
Introducing the Open Chain of Thought Leaderboard
Hugging Face has launched the Open Chain of Thought Leaderboard, a benchmarking platform specifically designed to evaluate open-weight language models on chain-of-thought reasoning capabilities. The leaderboard tracks model performance across reasoning-intensive tasks that require multi-step inference. This initiative aims to provide standardized, reproducible comparisons of CoT reasoning quality across the open-weights ecosystem.
Reasoning models struggle to control their chains of thought, and that's good
OpenAI introduces CoT-Control, a framework for evaluating how well reasoning models can deliberately manipulate or suppress their chain-of-thought outputs. The finding that models struggle to control their CoT is framed as a positive safety property, reinforcing the argument that visible reasoning traces serve as a meaningful monitorability safeguard. This contributes to ongoing research on whether chain-of-thought transparency is a reliable alignment and oversight tool.
Evaluating chain-of-thought monitorability
OpenAI introduces a framework and evaluation suite for assessing chain-of-thought monitorability, comprising 13 evaluations across 24 environments. The research finds that monitoring a model's internal reasoning is substantially more effective than monitoring outputs alone. The work is positioned as a step toward scalable oversight and control of increasingly capable AI systems.
Thinking with images
OpenAI announced a new capability allowing its reasoning models to incorporate images directly into their chain-of-thought process, enabling visual reasoning during intermediate thinking steps rather than only at input/output boundaries. This extends multimodal reasoning to the internal computation layer, potentially improving performance on tasks requiring visual analysis combined with multi-step reasoning. The announcement comes from OpenAI's official blog, indicating a product-level capability update.
Introducing OpenAI o1
OpenAI announced o1, a new series of AI models designed to spend more time 'thinking' before responding, using chain-of-thought reasoning to tackle complex problems in science, coding, and mathematics. The o1-preview and o1-mini models are being released, with o1-preview representing the most capable version and o1-mini offering a faster, cheaper alternative optimized for coding and reasoning tasks. OpenAI claims o1-preview ranks in the 89th percentile on competitive programming problems and performs at a PhD level on science benchmarks. This release marks a significant shift in OpenAI's approach to scaling, moving from purely training-time compute to inference-time compute as a new axis of capability improvement.
Learning to Reason with LLMs
OpenAI announced a new model or capability focused on reasoning in large language models, published on September 12, 2024. The post, hosted on the OpenAI blog, describes advances in training LLMs to perform complex multi-step reasoning. This likely corresponds to the release of the o1 (formerly 'Strawberry') model series, which uses chain-of-thought reasoning trained via reinforcement learning to achieve significantly improved performance on math, science, and coding benchmarks.
Research identifies 'commitment boundary' in chain-of-thought reasoning, enabling 55% CoT length reduction
A new arXiv preprint introduces the concept of a 'commitment boundary' in chain-of-thought reasoning — a sharp transition point where a model's answer stabilizes, after which subsequent reasoning steps are 'epiphenomenal' and causally inert. The authors use early-exit probing and attention probes to detect this boundary, finding it can be linearly decoded from intermediate steps and generalizes across tasks. Exploiting this signal to exit reasoning blocks at the commitment boundary reduces CoT length by up to 55% on average with negligible performance loss, with direct implications for inference efficiency in large reasoning models.
Improving Mathematical Reasoning with Process Supervision
OpenAI trained a model achieving state-of-the-art mathematical problem solving by rewarding each correct reasoning step (process supervision) rather than only the final answer (outcome supervision). This approach improves performance on math benchmarks and carries an alignment benefit by training models to produce human-endorsed chain-of-thought reasoning. The work highlights a potential synergy between capability improvements and alignment techniques.
Mistral AI Releases Magistral: First Reasoning Model in Open and Enterprise Variants
Mistral AI announces Magistral, its first reasoning model, released in two variants: Magistral Small (24B parameters, open-weight, Apache 2.0) and Magistral Medium (enterprise, closed). Magistral Medium scores 73.6% on AIME2024 (90% with majority voting @64), while Magistral Small scores 70.7% (83.3% respectively). Key differentiators include native multilingual chain-of-thought reasoning across eight major languages, transparent traceable reasoning steps, and up to 10x faster token throughput in Le Chat via Flash Answers. The release is accompanied by a research paper covering training infrastructure, reinforcement learning algorithm, and novel observations for training reasoning models.
Semantic vs. Surface Noise in LLM Agents: 68-Cell Measurement Study with Held-Out Validation
This paper documents an empirical phenomenon across 10 LLMs from 7 architecture families: meaning-bearing perturbations (paraphrase, synonym substitution) cause final-answer inconsistency ~19.69 percentage points more often than presentation-level perturbations (formatting, reordering) of comparable severity, across GSM8K, MATH, and HotpotQA benchmarks. The effect is validated on a held-out 11th model (qwen2.5-14B-Instruct) with 1,800 trajectories. Trace-level analysis supports a 'stealth-divergence' picture where semantic perturbations preserve the first action but induce divergence in intermediate reasoning steps, while two prior mechanism claims are explicitly retracted. The study is notable for its honest reporting of stress-test failures and pre-registered replication.
STORM: Internalized Spatial-Temporal Reasoning for Video-Language Models via Latent Trajectories
STORMS is a two-stage training framework that teaches large vision-language models to perform spatial-temporal video reasoning through bounded continuous latent trajectories rather than explicit textual chain-of-thought, keyframe selection, or external tool use. In Stage I, latent tokens are aligned with thought-video representations derived from generated videos; in Stage II, answer-only supervision internalizes the reasoning process. At inference time, no video regeneration or frame reinsertion is required, reducing latency and engineering complexity. Evaluations on VideoMME, MVBench, TempCompass, and MMVU show improved accuracy with substantially lower inference overhead versus tool-based pipelines.
Reasoning in Memory (RiM): Latent Reasoning via Working Memory Blocks in LLMs
RiM introduces a latent reasoning method that replaces autoregressive chain-of-thought token generation with fixed sequences of special 'memory block' tokens, allowing LLMs to perform internal computation without externalizing intermediate steps. These memory blocks are processed in a single forward pass rather than generated autoregressively, improving compute efficiency at test time. Training uses a two-stage curriculum: first grounding memory blocks by predicting explicit reasoning steps, then discarding step-level supervision and refining answers iteratively. Experiments across multiple model families and sizes show RiM matches or exceeds existing latent reasoning methods.
Question-Answering as Hidden State Probing for Test-Time Reasoning Intervention
This paper proposes using question-asking as an inference-time intervention to surface information about an LLM's hidden state during chain-of-thought reasoning. The authors train a probe on a student model's hidden states before and after question generation, finding it predictive of final answer correctness even before the teacher responds—suggesting self-diagnosis during question generation carries meaningful signal. They frame question-asking as a sequential decision problem with a gating policy, but find a gap between detection and recovery: interventions are as likely to harm correct trajectories as to fix incorrect ones. The results have implications for the limits of LLM self-refinement under uncertainty.
PolyGnosis 2.0: Multi-Agent Architecture for Prediction Market Intelligence via Harness Engineering
PolyGnosis 2.0 introduces a multi-agent system that synthesizes Polymarket prediction market signals with GDELT OSINT streams to identify 'Perspective Mismatches' as trading signals. The paper rigorously evaluates agentic harness engineering techniques—reflection loops, tool-calling, divide-and-conquer partitioning, and chain-of-thought—in high-noise financial domains. Key empirical findings include that structural partitioning is necessary for multi-dimensional alignment, but unconstrained terminal reflection induces logical drift, and a pervasive consensus bias emerges across agent configurations. The authors identify a Pareto-optimal configuration achieving professional-grade analytical precision with minimized latency and token overhead.
OneReason: Activating Chain-of-Thought Reasoning in Generative Recommendation Models
Researchers from the OneRec team introduce OneReason, a framework for enabling reasoning capabilities in generative recommendation models deployed across short-video, live-streaming, advertising, and e-commerce. The work identifies a key failure mode — that naive thinking-mode integration does not outperform non-thinking baselines — and diagnoses this as a deficit in two factors: itemic token perception and user behavior cognition. The proposed solution combines perception-focused pre-training, a three-level cognition-enhanced CoT format for supervised fine-tuning, and a specialize-then-unify RL training recipe.
PPC: Preplan-Plan-CoT Framework for LLM Mathematical Reasoning
This paper introduces PPC (Preplan-Plan-CoT), a reasoning framework that adds an explicit problem-understanding stage (the 'preplan') before the planning and chain-of-thought execution stages in LLM mathematical reasoning. The preplan captures problem type, applicable tools, and foreseeable pitfalls, addressing a gap in existing plan-based methods that only address 'how' to solve without first clarifying 'what' to solve. A three-stage synthesis pipeline with a spoiler-score detector and composite GRPO reward ensures clean preplan supervision and coherent plan generation. Evaluated across four backbones and five math benchmarks, PPC achieves best results on 39 of 40 metrics with +2.23 maj@16 and +3.06 pass@16 improvements over the strongest baseline at no additional inference token cost.
Qwen2.5-Math: Open-Source Mathematical LLM Series Released
Alibaba's Qwen team has released Qwen2.5-Math, an upgraded series of open-source mathematical LLMs including base and instruction-tuned models at 1.5B, 7B, and 72B parameter scales, plus a mathematical reward model. The models support Chain-of-Thought (CoT) and Tool-Integrated Reasoning (TIR) for English and Chinese math problem solving. This follows the Qwen2-Math release approximately one month prior and is claimed to be the leading open-source mathematical LLM series.
RREDCoT: Segment-level reward redistribution for chain-of-thought reasoning via self-approximated credit assignment
RREDCoT is a new method for redistributing rewards across segments of Chain-of-Thought traces during RL fine-tuning of reasoning language models, addressing the high-variance delayed-reward problem inherent in GRPO-style training. Rather than using computationally expensive Monte Carlo sampling for intermediate state value estimation, the method uses the model itself to approximate optimal reward redistribution without additional generation passes. The paper evaluates RREDCoT against MC sampling and several attribution baselines, analyzing segmentation strategies and state value estimation. This is relevant to the active research thread on improving RL fine-tuning stability and efficiency for reasoning models.
Dep-LLM: Training-free depression diagnosis framework using structured multi-factor LLM reasoning
Dep-LLM is a training-free framework for automatic depression detection from clinical interviews that uses frozen foundation LLMs without fine-tuning. The system decomposes long clinical dialogues into five thematic factors via Chain-of-Thought analysis, applies token-level entropy-based confidence modulation, and integrates multi-factor signals for final diagnosis. Evaluated on DAIC-WOZ and E-DAIC datasets, it outperforms zero-shot baselines across 21 foundation LLMs and surpasses supervised domain-specific and commercial LLMs on multiple metrics.