mlsanityData to 14 Oct 2026

Mi:dm K 2.5 Pro

arXiv 2603.18788 ↗First posted 19 Mar 2026cs.CL cs.AIcited by 0

On the map: LLM agents and tool use → General LLM agent frameworks and benchmarks

Abstract (arXiv)

The evolving LLM landscape requires capabilities beyond simple text generation, prioritizing multi-step reasoning, long-context understanding, and agentic workflows. This shift challenges existing models in enterprise environments, especially in Korean-language and domain-specific scenarios where scaling is insufficient. We introduce Mi:dm K 2.5 Pro, a 32B parameter flagship LLM designed to address enterprise-grade complexity through reasoning-focused optimization. Our methodology builds a robust data foundation via a quality-centric curation pipeline utilizing abstract syntax tree (AST) analysis for code, gap-filling synthesis for mathematics, and an LLM-based quality evaluator. Pre-training scales the model via layer-predictor-based Depth Upscaling (DuS) and a progressive strategy supporting a 128K token context window. Post-training introduces a specialized multi-stage pipeline, including Reasoning SFT, model merging, and asynchronous reinforcement learning (RL), to develop complex problem-solving skills. "Fusion Training" then rebalances these capabilities with conversational fluency, consistent response styling, and reliable tool-use. The evaluations show that Mi:dm K 2.5 Pro achieves competitive performance against leading global and domestic models. In addition, it sets state-of-the-art results on Korean-specific benchmarks, showcasing deep linguistic and cultural understanding. Finally, Responsible AI evaluations validate safety against attacks, ensuring a secure profile for deployment with a balance of harmlessness and responsiveness.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in General LLM agent frameworks and benchmarks, one of 16 directions in LLM agents and tool use. Builds on 70 of its 88 references that are in the map. First posted on arXiv 19 Mar 2026.

Builds on

70 of 88 references.

Closest ancestors · 1

Same direction, newest first

  1. 2024τ-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains17 Jun

Other references · 69

  1. 2017Proximal Policy Optimization Algorithmscited by 15,215
  2. 2021Training Verifiers to Solve Math Word Problemscited by 10,109
  3. 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,944
  4. 2021Measuring Mathematical Problem Solving With the MATH Datasetcited by 5,371
  5. 2022Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedbackcited by 3,720
  6. 2023GPQA: A Graduate-Level Google-Proof Q&A Benchmarkcited by 2,879
  7. 2022Self-Instruct: Aligning Language Models with Self-Generated Instructionscited by 2,724
  8. 2022Emergent Abilities of Large Language Modelscited by 2,512
  9. 2023G-Eval: NLG Evaluation using GPT-4 with Better Human Alignmentcited by 2,276
  10. 2022Survey of Hallucination in Natural Language Generationcited by 2,075
  11. 2024HybridFlow: A Flexible and Efficient RLHF Frameworkcited by 2,057
  12. 2024LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Codecited by 2,004
38 more
  1. 2024MMLU-Pro: A More Robust and Challenging Multi-Task Language Understanding Benchmarkcited by 1,848
  2. 2022Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference timecited by 1,396
  3. 2023Is Your Code Generated by ChatGPT Really Correct? Rigorous Evaluation of Large Language Models for Code Generationcited by 1,391
  4. 2025gpt-oss-120b & gpt-oss-20b Model Cardcited by 1,386
  5. 2024RULER: What's the Real Context Size of Your Long-Context Language Models?cited by 1,346
  6. 2023Instruction-Following Evaluation for Large Language Modelscited by 1,228
  7. 2021A General Language Assistant as a Laboratory for Alignmentcited by 1,055
  8. 2018IMPALA: Scalable Distributed Deep-RL with Importance Weighted Actor-Learner Architecturescited by 883
  9. 2023The Flan Collection: Designing Data and Methods for Effective Instruction Tuningcited by 739
  10. 2024MiniCPM: Unveiling the Potential of Small Language Models with Scalable Training Strategiescited by 723
  11. 2024Yi: Open Foundation Models by 01.AIcited by 719
  12. 2025τ2-Bench: Evaluating Conversational Agents in a Dual-Control Environmentcited by 627
  13. 2023YaRN: Efficient Context Window Extension of Large Language Modelscited by 601
  14. 2026Terminal-Bench: Benchmarking Agents on Hard, Realistic Tasks in Command Line Interfacescited by 585
  15. 2024Scaling Synthetic Data Creation with 1,000,000,000 Personascited by 417
  16. 2024Molmo and PixMo: Open Weights and Open Data for State-of-the-Art Vision-Language Modelscited by 329
  17. 2025Model Context Protocol (MCP): Landscape, Security Threats, and Future Research Directionscited by 265
  18. 2021Curriculum Learning: A Surveycited by 259
  19. 2024MT-Bench-101: A Fine-Grained Benchmark for Evaluating Large Language Models in Multi-Turn Dialoguescited by 250
  20. 2025AReaL: A Large-Scale Asynchronous Reinforcement Learning System for Language Reasoningcited by 250
  21. 2024RegMix: Data Mixture as Regression for Language Model Pre-trainingcited by 177
  22. 2023SOLAR 10.7B: Scaling Large Language Models with Simple yet Effective Depth Up-Scalingcited by 172
  23. 2025Qwen2.5-1M Technical Reportcited by 164
  24. 2024Tokenization counts: the impact of tokenization on arithmetic in frontier LLMscited by 115
  25. 2024KMMLU: Measuring Massive Multitask Language Understanding in Koreancited by 91
  26. 2024MT-Eval: A Multi-Turn Capabilities Evaluation Benchmark for Large Language Modelscited by 81
  27. 2023KoBBQ: Korean Bias Benchmark for Question Answeringcited by 71
  28. 2024CLIcK: A Benchmark Dataset of Cultural and Linguistic Intelligence in Koreancited by 66
  29. 2025Command A: An Enterprise-Ready Large Language Modelcited by 59
  30. 2024WARP: On the Benefits of Weight Averaged Rewarded Policiescited by 45
  31. 2025Understand, Solve and Translate: Bridging the Multilingual Mathematical Reasoning Gapcited by 43
  32. 2025When2Call: When (not) to Call Toolscited by 42
  33. 2025AsyncFlow: An Asynchronous Streaming RL Framework for Efficient LLM Post-Trainingcited by 41
  34. 2025Model Merging in Pre-training of Large Language Modelscited by 37
  35. 2025Enhancing Multilingual LLM Pretraining with Model-Based Data Selectioncited by 35
  36. 2025ToolDial: Multi-turn Dialogue Generation Method for Tool-Augmented Language Modelscited by 34
  37. 2025RL Squeezes, SFT Expands: A Comparative Study of Reasoning LLMscited by 34
  38. 2024KorNAT: LLM Alignment Benchmark for Korean Social Values and Common Knowledgecited by 33

50 most cited of 69 All 69 (JSON)

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2024Inference Scaling vs Reasoning: An Empirical Analysis of Compute-Optimal LLM Problem-Solving
  2. 2026The Periodic Table of LLM Reasoning: A Structured Survey of Reasoning Paradigms, Methods, and Failure Modes
  3. 2025Scaling LLM Planning: NL2FLOW for Parametric Problem Generation and Rigorous Evaluation
  4. 2025How Difficulty-Aware Staged Reinforcement Learning Enhances LLMs' Reasoning Capabilities: A Preliminary Experimental Study
  5. 2023Have LLMs Advanced Enough? A Challenging Problem Solving Benchmark For Large Language Models
  6. 2023Competition-Level Problems are Effective LLM Evaluators
  7. 2024LLM2: Let Large Language Models Harness System 2 Reasoning
  8. 2024Navigating Complexity: Orchestrated Problem Solving with Multi-Agent LLMs