Pith. sign in

REVIEW 19 cited by

MiMo: Unlocking the Reasoning Potential of Language Model -- From Pretraining to Posttraining

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 2505.07608 v2 pith:H37IGTXU submitted 2025-05-12 cs.CL cs.AIcs.LG

classification cs.CLcs.AIcs.LG
keywords modelreasoningdataperformancepotentialduringlanguagemathematics
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

We present MiMo-7B, a large language model born for reasoning tasks, with optimization across both pre-training and post-training stages. During pre-training, we enhance the data preprocessing pipeline and employ a three-stage data mixing strategy to strengthen the base model's reasoning potential. MiMo-7B-Base is pre-trained on 25 trillion tokens, with additional Multi-Token Prediction objective for enhanced performance and accelerated inference speed. During post-training, we curate a dataset of 130K verifiable mathematics and programming problems for reinforcement learning, integrating a test-difficulty-driven code-reward scheme to alleviate sparse-reward issues and employing strategic data resampling to stabilize training. Extensive evaluations show that MiMo-7B-Base possesses exceptional reasoning potential, outperforming even much larger 32B models. The final RL-tuned model, MiMo-7B-RL, achieves superior performance on mathematics, code and general reasoning tasks, surpassing the performance of OpenAI o1-mini. The model checkpoints are available at https://github.com/xiaomimimo/MiMo.

Discussion (0). Sign in to comment.

Forward citations

Cited by 19 Pith papers

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

  1. rStar2-Agent: Agentic Reasoning Technical Report

    cs.CL 2025-08 conditional novelty 7.0 of 10

    A 14B model trained with agentic RL and a resample-on-correct rollout strategy scores 80.6% on AIME24 and 69.8% on AIME25, nearly matching DeepSeek-R1 (671B) in one week on 64 GPUs.

  2. Behavioural Signatures of Risk-Sensitive Decision-Making in Large Language Models

    cs.AI 2026-07 conditional novelty 6.5 of 10

    LLMs display stable Participation and Proactiveness risk profiles in multi-agent poker that remain largely robust to opponent mix and adapt heterogeneously under global and personal risk pressure.

  3. OralAgent: Integrating Reasoning, Tools, and Knowledge for Interactive Dental Image Analysis

    cs.CL 2026-04 accept novelty 6.5 of 10

    OralAgent, a ReAct-style dental agent with 22 vision tools and a 134.8M-token textbook RAG corpus, reaches SOTA on MMOral-Uni, MMOral-OPG, and OralQA-ZH.

  4. CodePercept: Code-Grounded Visual STEM Perception for MLLMs

    cs.CV 2026-03 conditional novelty 6.5 of 10

    Perception, not reasoning, is the main bottleneck for MLLM STEM visual reasoning, and training on executable reconstruction code measurably fixes it.

  5. PIVOT: Efficient Query-Group Indexing for Token-Level Sparse Attention

    cs.CL 2026-07 conditional novelty 6.0 of 10

    Sharing one proxy full-prefix scan across nearby query groups matches dense DSA indexer accuracy while accelerating indexing up to 4× and end-to-end latency up to 1.6×.

  6. RefBench-PRO: Perceptual and Reasoning Oriented Benchmark for Referring Expression Comprehension

    cs.CV 2025-12 conditional novelty 6.0 of 10

    RefBench-PRO organizes REC into attribute, position, interaction, relation, commonsense, and reject tasks; no tested MLLM exceeds 72%, and Ref-R1 raises Qwen2.5-VL-7B from 57.6 to 69.4 on it.

  7. Efficiency of turbulence

    physics.flu-dyn 2025-08 unverdicted novelty 6.0 of 10

    The efficiency of turbulence, the fraction of input energy stored in the flow, appears bounded and may saturate in a power-law manner across several turbulent flows.

  8. Are Reasoning Models More Prone to Hallucination?

    cs.CL 2025-05 conditional novelty 6.0 of 10

    Post-training pipeline choice (SFT+RL vs RL-only vs SFT-only) reliably shifts hallucination rates in large reasoning models on fact-seeking benchmarks.

  9. X$^3$-OPD: Distilling Reasoning into Large Audio-Language Models via On-Policy Alignment

    cs.LG 2026-07 conditional novelty 5.0 of 10

    X3-OPD improves audio-grounded reasoning by training the audio student on its own rollouts with token-level teacher feedback, using a three-tier paired text-audio corpus.

  10. Self-Improving is Often Sudden: Enlightenment-style Finetuning for Large-Scale Models

    cs.LG 2026-07 conditional novelty 5.0 of 10

    A training-free intervention that scales VLM residual connections and mixes LLM attention heads claims 1–3 point zero-shot accuracy gains, but the evidence is partly tuned on the reported benchmarks.

  11. Enhancing Large Multimodal Models in Key Information Extraction via Scene-Aware Document Synthesis

    cs.CV 2026-07 conditional novelty 5.0 of 10

    Scene-aware multi-agent document synthesis plus error-driven hard-example expansion improves compact Qwen3-VL models on constrained and open-category KIE, topping reported on-device baselines.

  12. DSBench: A Comprehensive Benchmark for Evaluating External and In-Cabin Risks

    cs.RO 2025-11 reject novelty 5.0 of 10

    A new benchmark claims to be the first to test VLMs on both external and in-cabin driving risks, and reports a fine-tuned model far outperforming all baselines.

  13. AceReason-Nemotron 1.1: Advancing Math and Code Reasoning through SFT and RL Synergy

    cs.CL 2025-06 conditional novelty 5.0 of 10

    A 7B reasoning model trained with carefully balanced SFT and RL beats prior small models on math and code benchmarks, with the paper documenting scaling and temperature heuristics.

  14. MiniCPM4: Ultra-Efficient LLMs on End Devices

    cs.CL 2025-06 conditional novelty 5.0 of 10

    MiniCPM4-8B reportedly matches Qwen3-8B on standard benchmarks while using about 22% of the training tokens, and achieves large long-context speedups on edge devices.

  15. MiMo-VL Technical Report

    cs.CL 2025-06 conditional novelty 5.0 of 10

    MiMo-VL-7B-RL, a 7B open-source vision-language model, reports state-of-the-art results on 35 of 40 benchmarks and a 59.4 OlympiadBench score, with the report crediting long-CoT pretraining data and mixed on-policy RL.

  16. Unleashing the Reasoning Potential of Pre-trained LLMs by Critique Fine-Tuning on One Problem

    cs.CL 2025-06 conditional novelty 5.0 of 10

    One-shot critique fine-tuning, training on critiques of candidate solutions to a single problem, yields large reasoning gains on math and logic benchmarks at far lower compute than one-shot RL.

  17. General-Reasoner: Advancing LLM Reasoning Across All Domains

    cs.CL 2025-05 conditional novelty 5.0 of 10

    Zero-RL training with a compact generative answer verifier and diverse verifiable questions improves LLM reasoning on general-domain benchmarks while retaining mathematical ability.

  18. Wait, We Don't Need to "Wait"! Removing Thinking Tokens Improves Reasoning Efficiency

    cs.CL 2025-06 conditional novelty 4.0 of 10

    Suppressing "Wait"-like reflection tokens at decode time reduces reasoning token counts by 27-51% across five R1-style model families, with mixed accuracy effects.

  19. Towards Concise and Adaptive Thinking in Large Reasoning Models: A Survey

    cs.AI 2025-07 conditional novelty 3.0 of 10

    A comprehensive review that categorizes methods for shortening and adaptively triggering chain-of-thought reasoning in large language models.

Pith tools