mlsanityData to 7 Oct 2026

DeepSeek-V3.2: Pushing the Frontier of Open Large Language Models

arXiv 2512.02556 ↗First posted 2 Dec 2025cs.CLcited by 755

Not on the map. Why?

Cited by 248 in LLM agents and tool use, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

We introduce DeepSeek-V3.2, a model that harmonizes high computational efficiency with superior reasoning and agent performance. The key technical breakthroughs of DeepSeek-V3.2 are as follows: (1) DeepSeek Sparse Attention (DSA): We introduce DSA, an efficient attention mechanism that substantially reduces computational complexity while preserving model performance in long-context scenarios. (2) Scalable Reinforcement Learning Framework: By implementing a robust reinforcement learning protocol and scaling post-training compute, DeepSeek-V3.2 performs comparably to GPT-5. Notably, our high-compute variant, DeepSeek-V3.2-Speciale, surpasses GPT-5 and exhibits reasoning proficiency on par with Gemini-3.0-Pro, achieving gold-medal performance in both the 2025 International Mathematical Olympiad (IMO) and the International Olympiad in Informatics (IOI). (3) Large-Scale Agentic Task Synthesis Pipeline: To integrate reasoning into tool-use scenarios, we developed a novel synthesis pipeline that systematically generates training data at scale. This methodology facilitates scalable agentic post-training, yielding substantial improvements in generalization and instruction-following robustness within complex, interactive environments.

Key facts. Cited by 755 papers in the map (data through 7 Oct 2026). Builds on 17 of its 24 references that are in the map. First posted on arXiv 2 Dec 2025.

Builds on

17 of 24 references.

References · 17

  1. 2017Attention Is All You Needcited by 59,932
  2. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 8,851
  3. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,146
  4. 2025Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilitiescited by 3,655
  5. 2023GPQA: A Graduate-Level Google-Proof Q&A Benchmarkcited by 2,748
  6. 2024MMLU-Pro: A More Robust and Challenging Multi-Task Language Understanding Benchmarkcited by 1,758
  7. 2024DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Modelcited by 1,093
  8. 2019Fast Transformer Decoding: One Write-Head is All You Needcited by 700
  9. 2025BrowseComp: A Simple Yet Challenging Benchmark for Browsing Agentscited by 616
  10. 2025τ2-Bench: Evaluating Conversational Agents in a Dual-Control Environmentcited by 558
  11. 2025Humanity's Last Examcited by 467
  12. 2025GLM-4.5: Agentic, Reasoning, and Coding (ARC) Foundation Modelscited by 446
5 more
  1. 2025Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attentioncited by 391
  2. 2025MathArena: Evaluating LLMs on Uncontaminated Math Competitionscited by 275
  3. 2025MCP-Universe: Benchmarking Large Language Models with Real-World Model Context Protocol Serverscited by 89
  4. 2025The Tool Decathlon: Benchmarking Language Agents for Diverse, Realistic, and Long-Horizon Task Executioncited by 73
  5. 2025Towards Robust Mathematical Reasoningcited by 67

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 20; 2026: 735.

755 papers.

Sort:

By area:

  1. 2026GLM-5: from Vibe Coding to Agentic Engineeringcited by 502
  2. 2026Kimi K2.5: Visual Agentic Intelligencecited by 449
  3. 2026GDPO: Group reward-Decoupled Normalization Policy Optimization for Multi-reward RL Optimizationcited by 175
  4. 2026MiMo-V2-Flash Technical Reportcited by 163
  5. 2026SkillX: Automatically Constructing Skill Knowledge Bases for Agentscited by 69
  6. 2026DeepSeek-OCR 2: Visual Causal Flowcited by 68
  7. 2026Qwen3-Coder-Next Technical Reportcited by 60
  8. 2026Claw-Eval: Towards Trustworthy Evaluation of Autonomous Agentscited by 57
  9. 2025NL2Repo-Bench: Towards Long-Horizon Repository Generation Evaluation of Coding Agentscited by 56
  10. 2026Agent World Model: Infinity Synthetic Environments for Agentic Reinforcement Learningcited by 51
  11. 2025SWE-EVO: Benchmarking Coding Agents in Long-Horizon Software Evolution Scenarioscited by 49
  12. 2025Nemotron-Cascade: Scaling Cascaded Reinforcement Learning for General-Purpose Reasoning Modelscited by 45
38 more
  1. 2026REDSearcher: A Scalable and Cost-Efficient Framework for Long-Horizon Search Agentscited by 43
  2. 2025Toward Training Superintelligent Software Agents through Self-Play SWE-RLcited by 37
  3. 2026Agent-World: Scaling Real-World Environment Synthesis for Evolving General Agent Intelligencecited by 35
  4. 2026On Data Engineering for Scaling LLM Terminal Capabilitiescited by 34
  5. 2026Nemotron 3 Super: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoningcited by 33
  6. 2026Rethinking the Trust Region in LLM Reinforcement Learningcited by 32
  7. 2026OpenSeeker: Democratizing Frontier Search Agents by Fully Open-Sourcing Training Datacited by 32
  8. 2026AOrchestra: Automating Sub-Agent Creation for Agentic Orchestrationcited by 30
  9. 2026WildClawBench: A Benchmark for Real-World, Long-Horizon Agent Evaluationcited by 29
  10. 2026Unsafer in Many Turns: Benchmarking and Defending Multi-Turn Safety Risks in Tool-Using Agentscited by 28
  11. 2026SkillCraft: Can LLM Agents Learn to Use Tools Skillfully?cited by 28
  12. 2026AgentProcessBench: Diagnosing Step-Level Process Quality in Tool-Using Agentscited by 28
  13. 2026Large-Scale Terminal Agentic Trajectory Generation from Dockerized Environmentscited by 27
  14. 2026IndexCache: Accelerating Sparse Attention via Cross-Layer Index Reusecited by 27
  15. 2026Toward Ultra-Long-Horizon Agentic Science: Cognitive Accumulation for Machine Learning Engineeringcited by 26
  16. 2026SpecBench: Measuring Reward Hacking in Long-Horizon Coding Agentscited by 25
  17. 2026InternAgent-1.5: A Unified Agentic Framework for Long-Horizon Autonomous Scientific Discoverycited by 24
  18. 2026Benchmark Test-Time Scaling of General LLM Agentscited by 24
  19. 2026ScaleEnv: Scaling Environment Synthesis from Scratch for Generalist Interactive Tool-Use Agent Trainingcited by 23
  20. 2026Beyond Trajectory-Level Attribution: Graph-Based Credit Assignment for Agentic Reinforcement Learningcited by 23
  21. 2026ReCreate: Reasoning and Creating Domain Agents Driven by Experiencecited by 22
  22. 2026WorldCompass: Reinforcement Learning for Long-Horizon World Modelscited by 22
  23. 2026SLA2: Sparse-Linear Attention with Learnable Routing and QATcited by 21
  24. 2026OmniGAIA: Towards Native Omni-Modal AI Agentscited by 21
  25. 2026Immersion in the GitHub Universe: Scaling Coding Agents to Masterycited by 20
  26. 2026BeyondSWE: Can Current Code Agent Survive Beyond Single-Repo Bug Fixing?cited by 20
  27. 2026Composer 2 Technical Reportcited by 20
  28. 2026On the Design of Qwen3.8-Next Architecture: Evaluation, Efficiency, and Training Stabilitycited by 20
  29. 2025FrontierCS: Evolving Challenges for Evolving Intelligencecited by 19
  30. 2026AgentNoiseBench: Benchmarking Robustness of Tool-Using LLM Agents Under Noisy Conditioncited by 19
  31. 2026OpenRT: An Open-Source Red Teaming Framework for Multimodal LLMscited by 18
  32. 2026SWE-Master: Unleashing the Potential of Software Engineering Agents via Post-Trainingcited by 18
  33. 2026Agentic Aggregation for Parallel Scaling of Long-Horizon Agentic Taskscited by 18
  34. 2026LongBench Pro: A More Realistic and Comprehensive Bilingual Long-Context Evaluation Benchmarkcited by 17
  35. 2026From Self-Evolving Synthetic Data to Verifiable-Reward RL: Post-Training Multi-turn Interactive Tool-Using Agentscited by 17
  36. 2026SWE-Universe: Scale Real-World Verifiable Environments to Millionscited by 16
  37. 2026SpargeAttention2: Trainable Sparse Attention via Hybrid Top-k+Top-p Masking and Distillation Fine-Tuningcited by 16
  38. 2026Scaling the Horizon, Not the Parameters: Reaching Trillion-Parameter Performance with a 35B Agentcited by 16

50 most cited of 755 All 755 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2025A Review of DeepSeek Models' Key Innovative Techniques
  2. 2025Insights into DeepSeek-V3: Scaling Challenges and Reflections on Hardware for AI Architectures
  3. 2025DeepSeek: Paradigm Shifts and Technical Evolution in Large AI Models
  4. 2026DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence
  5. 2025100 Days After DeepSeek-R1: A Survey on Replication Studies and More Directions for Reasoning Language Models
  6. 2024DeepSeek-V3 Technical Report
  7. 2024DeepSeek LLM: Scaling Open-Source Language Models with Longtermism
  8. 2025Benchmarking Reasoning Robustness in Large Language Models