Pith. sign in

REVIEW 23 cited by

LlamaV-o1: Rethinking Step-by-step Visual Reasoning in LLMs

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2501.06186 v1 pith:UB2L5KJT submitted 2025-01-10 cs.CV

classification cs.CV
keywords reasoningvisualllamav-o1multi-stepbenchmarkmodelsstep-by-stepsteps
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Reasoning is a fundamental capability for solving complex multi-step problems, particularly in visual contexts where sequential step-wise understanding is essential. Existing approaches lack a comprehensive framework for evaluating visual reasoning and do not emphasize step-wise problem-solving. To this end, we propose a comprehensive framework for advancing step-by-step visual reasoning in large language models (LMMs) through three key contributions. First, we introduce a visual reasoning benchmark specifically designed to evaluate multi-step reasoning tasks. The benchmark presents a diverse set of challenges with eight different categories ranging from complex visual perception to scientific reasoning with over 4k reasoning steps in total, enabling robust evaluation of LLMs' abilities to perform accurate and interpretable visual reasoning across multiple steps. Second, we propose a novel metric that assesses visual reasoning quality at the granularity of individual steps, emphasizing both correctness and logical coherence. The proposed metric offers deeper insights into reasoning performance compared to traditional end-task accuracy metrics. Third, we present a new multimodal visual reasoning model, named LlamaV-o1, trained using a multi-step curriculum learning approach, where tasks are progressively organized to facilitate incremental skill acquisition and problem-solving. The proposed LlamaV-o1 is designed for multi-step reasoning and learns step-by-step through a structured training paradigm. Extensive experiments show that our LlamaV-o1 outperforms existing open-source models and performs favorably against close-source proprietary models. Compared to the recent Llava-CoT, our LlamaV-o1 achieves an average score of 67.3 with an absolute gain of 3.8\% across six benchmarks while being 5 times faster during inference scaling. Our benchmark, model, and code are publicly available.

Discussion (0). Sign in to comment.

Forward citations

Cited by 23 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Evaluating MLLMs with Multimodal Multi-image Reasoning Benchmark

    cs.CV 2025-06 conditional novelty 7.0 of 10

    MMRB is the first benchmark combining multi-image inputs with chain-of-thought reasoning annotations, and its evaluation shows open-source MLLMs trail commercial models while multi-image reward models are unstable.

  2. More Thinking, Less Seeing? Assessing Amplified Hallucination in Multimodal Reasoning Models

    cs.CL 2025-05 conditional novelty 7.0 of 10

    Reasoning models trade visual grounding for language-based inference, and this paper measures that trade-off with a new metric and benchmark.

  3. Towards Spoken Mathematical Reasoning: Benchmarking Speech-based Models over Multi-faceted Math Problems

    cs.CL 2025-05 conditional novelty 7.0 of 10

    A new spoken math benchmark, Spoken-MQA, shows that current speech-based AI models reason poorly from spoken math input, especially for arithmetic and knowledge-heavy problems.

  4. SyncLoop: A Multimodal Dual-Loop Framework for Self-Improving Mathematical Reasoning

    cs.CV 2025-07 conditional novelty 6.0 of 10

    SyncLoop jointly evolves multimodal training data and model capability through alternating SFT and RL, selecting error-prone samples to improve geometry reasoning.

  5. Bottom-up Domain-specific Superintelligence: A Reliable Knowledge Graph is What We Need

    cs.CL 2025-07 conditional novelty 6.0 of 10

    A language model fine-tuned on knowledge-graph-path reasoning tasks (QwQ-Med-3) beats strong baselines on a same-style benchmark but shows mixed gains on external medical QA tests.

  6. VisionThink: Smart and Efficient Vision Language Model via Reinforcement Learning

    cs.CV 2025-07 conditional novelty 6.0 of 10

    A vision-language model learns via reinforcement learning when to upscale a low-resolution image, cutting visual tokens roughly in half while preserving accuracy on most benchmarks.

  7. MagiC: Evaluating Multimodal Cognition Toward Grounded Visual Reasoning

    cs.CV 2025-07 conditional novelty 6.0 of 10

    MagiC evaluates answer correctness, reasoning validity, grounding fidelity, and self-correction on about 900 hand-annotated visual questions across 15 vision-language models.

  8. Ascending the Infinite Ladder: Benchmarking Spatial Deformation Reasoning in Vision-Language Models

    cs.CV 2025-07 conditional novelty 6.0 of 10

    Vision-language models can handle some 2D shape puzzles but nearly all fail at multi-step 3D spatial deformation reasoning.

  9. Ego-R1: Chain-of-Tool-Thought for Ultra-Long Egocentric Video Reasoning

    cs.CV 2025-06 conditional novelty 6.0 of 10

    A 3B agent trained with supervised tool-use traces and reinforcement learning answers week-long egocentric video questions by dynamically selecting hierarchical retrieval, video-LLM, and VLM tools.

  10. VRBench: A Benchmark for Multi-Step Reasoning in Long Narrative Videos

    cs.CV 2025-06 conditional novelty 6.0 of 10

    VRBench is a benchmark of 960 long narrative videos with 8,243 human-written multi-step questions, plus a two-level evaluation of answer accuracy and reasoning quality for 31 large models.

  11. Rex-Thinker: Grounded Object Referring via Chain-of-Thought Reasoning

    cs.CV 2025-06 conditional novelty 6.0 of 10

    Training a multimodal LLM on GPT-4o-generated chain-of-thought referring traces, then optimizing with GRPO, improves referring accuracy and abstention on HumanRef.

  12. VideoCap-R1: Enhancing MLLMs for Video Captioning via Structured Thinking

    cs.CV 2025-06 conditional novelty 6.0 of 10

    GRPO post-training with structured thinking and dual think/caption rewards improves Qwen2-VL-7B video captioning over the base model and SFT on DREAM-1K, VDC, and CAREBENCH.

  13. GThinker: Towards General Multimodal Reasoning via Cue-Guided Rethinking

    cs.CV 2025-06 conditional novelty 6.0 of 10

    A cue-tagging and rethinking training recipe lifts a 7B multimodal model to 81.5% on M3CoT, though the gain may reflect training on the same benchmark.

  14. Agent-X: Evaluating Deep Multimodal Reasoning in Vision-Centric Agentic Tasks

    cs.CV 2025-05 conditional novelty 6.0 of 10

    Agent-X, a new 828-task benchmark with step-level scoring, finds that the best multimodal agents still fail over half of real-world visual tool-use tasks.

  15. MME-Reasoning: A Comprehensive Benchmark for Logical Reasoning in MLLMs

    cs.AI 2025-05 conditional novelty 6.0 of 10

    A new 1,188-question multimodal benchmark covering deductive, inductive, and abductive reasoning shows that leading MLLMs score around 60% and are especially weak at abductive reasoning.

  16. Vad-R1: Towards Video Anomaly Reasoning via Perception-to-Cognition Chain-of-Thought

    cs.CV 2025-05 conditional novelty 6.0 of 10

    A two-stage trained MLLM with a perception-to-cognition chain-of-thought and a self-verification RL reward outperforms prior models on video anomaly detection and reasoning.

  17. ARB: A Comprehensive Arabic Multimodal Reasoning Benchmark

    cs.CV 2025-05 conditional novelty 6.0 of 10

    ARB provides 1,356 Arabic multimodal questions with 5,119 human-reviewed reasoning steps and shows leading models score much higher on reasoning fluency than on correct answers.

  18. WeThink: Toward General-purpose Vision-Language Reasoning via Reinforcement Learning

    cs.CV 2025-06 conditional novelty 5.0 of 10

    WeThink, a 120K-image QA dataset with AI-generated reasoning paths, combined with hybrid-reward reinforcement learning, improves a 7B vision-language model across 14 benchmarks.

  19. From Street Views to Urban Science: Discovering Road Safety Factors with Multimodal Large Language Models

    cs.LG 2025-06 reject novelty 5.0 of 10

    UrbanX uses LLM-generated visual questions, MLLM answers, and linear regression to predict crash rates, claiming better performance than ResNet/ViT while keeping features interpretable.

  20. Advancing Multimodal Reasoning via Reinforcement Learning with Cold Start

    cs.CL 2025-05 conditional novelty 5.0 of 10

    Using a two-stage SFT-plus-GRPO pipeline, the authors train 3B and 7B multimodal models that beat prior open-source models on four math reasoning benchmarks, while showing that reflective 'aha moment' text is not a re...

  21. See, Think, Learn: A Self-Taught Multimodal Reasoner

    cs.CV 2025-12 conditional novelty 4.0 of 10

    A self-training loop that structures model rationales into image-caption, reasoning, and conclusion, and adds negative rationales, improves VLM accuracy on M3CoT over answer-only and STaR baselines.

  22. A Comprehensive Survey on Trustworthiness in Reasoning with Large Language Models

    cs.CL 2025-09 conditional novelty 4.0 of 10

    A structured literature survey concluding that reasoning capabilities do not automatically make LLMs more trustworthy and can introduce new vulnerabilities in safety, robustness, and privacy.

  23. Enhancing Scientific Visual Question Answering through Multimodal Reasoning and Ensemble Modeling

    cs.CV 2025-07 conditional novelty 4.0 of 10

    On the SciVQA 2025 benchmark, an InternVL3 model with optimized prompts and chain-of-thought instructions reaches ROUGE-1 and ROUGE-L F1 of 0.740, and a figure-type-aware ensemble ranks 5th.

Pith tools