模型即裁判论文趋势 2023–2026

自动评测

统计范围：NeurIPS、ICLR、ICML、CVPR 四个会议，2023–2026 年，共 401 篇被归入该主题的论文。

主要结论

每年论文数

各会议论文数

按年份和会议统计

模型即裁判论文数，按年份和会议
年份NeurIPSICLRICMLCVPR合计
202351208
202427129048
20255337333126
20268862645219

代表论文

按年份和会议均匀挑选，方便快速看到每一届的样貌

全部模型即裁判论文

年份全部2023202420252026
会议全部NeurIPSICLRICMLCVPR

共 219 篇，第 2 / 5 页

  1. ICLR 2026Rethinking LLM-as-a-Judge: Representation-as-a-Judge with Small Language Models via Semantic Capacity Asymmetry
  2. ICLR 2026RouterArena: An Open Platform for Comprehensive Comparison of LLM Routers
  3. ICLR 2026SCI-Verifier: Scientific Verifier with Thinking
  4. ICLR 2026Semantic Voting: A Self-Evaluation-Free Approach for Efficient LLM Self-Improvement on Unverifiable Open-ended Tasks
  5. ICLR 2026SK2Decompile: LLM-based Two-Phase Binary Decompilation from Skeleton to Skin
  6. ICLR 2026SocialHarmBench: Revealing LLM Vulnerabilities to Socially Harmful Requests
  7. ICLR 2026SparseEval: Efficient Evaluation of Large Language Models by Sparse Optimization
  8. ICLR 2026STAR: Strategy-driven Automatic Jailbreak Red-teaming For Large Language Model
  9. ICLR 2026Talk, Evaluate, Diagnose: User-aware Agent Evaluation with Automated Error Analysis
  10. ICLR 2026The Ideation-Execution Gap: Execution Outcomes of LLM-Generated versus Human Research Ideas
  11. ICLR 2026The Unseen Frontier: Pushing the Limits of LLM Sparsity with Surrogate-Free ADMM
  12. ICLR 2026Towards Reliable Benchmarking: A Contamination Free, Controllable Evaluation Framework for Multi-step LLM Function Calling
  13. ICLR 2026TrustJudge: Inconsistencies of LLM-as-a-Judge and How to Alleviate Them
  14. ICLR 2026Unlocking Full Efficiency of Token Filtering in Large Language Model Training
  15. ICLR 2026Watermarking Diffusion Language Models
  16. ICLR 2026WebDevJudge: Evaluating (M)LLMs as Critiques for Web Development Quality
  17. ICLR 2026Your Agent May Misevolve: Emergent Risks in Self-evolving LLM Agents
  18. ICML 2026A Coin Flip for Safety: LLM Judges Fail to Reliably Measure Adversarial Robustness
  19. ICML 2026A Judge-Aware Ranking Framework for Evaluating Large Language Models without Ground Truth
  20. ICML 2026Adaptive Contracts for Cost-Effective AI Delegation
  21. ICML 2026Are LLM Evaluators Really Narcissists? Sanity Checking Self-Preference Evaluations
  22. ICML 2026Automatic Construction of Clinical Scoring Systems with LLM Agents
  23. ICML 2026Beyond Drift: Stabilizing Subjective LLM Evaluation with Information-Theoretic Rubrics
  24. ICML 2026Breaking the Exploration Bottleneck: Rubric-Scaffolded Reinforcement Learning for Open-Ended LLM Reasoning
  25. ICML 2026CARE: Confounder-Aware Aggregation for Reliable LLM Evaluation
  26. ICML 2026Controlled LLM Training on Spectral Sphere
  27. ICML 2026Cutting LLM Evaluation Costs with SySRs: A Bandit Algorithm That Provably Exploits Model Similarity
  28. ICML 2026D-Judge: Disrupting Multi-Turn Jailbreaks using Semantics-Preserving Output Rewriting
  29. ICML 2026Demystifying LLM-as-a-Judge: Analytically Tractable Model for Inference-Time Scaling
  30. ICML 2026Dependence-Aware Label Aggregation for LLM-as-a-Judge via Ising Models
  31. ICML 2026Diagnosing the Reliability of LLM-as-a-Judge via Item Response Theory
  32. ICML 2026Distribution-Calibrated Inference Time Compute for Thinking LLM-as-a-Judge
  33. ICML 2026Distributional Open-Ended Evaluation of LLM Cultural Value Alignment Based on Value Codebook
  34. ICML 2026Does Reinforcement Fine-Tuning Improve Generalization of LLM Agents? An Empirical Study
  35. ICML 2026Easier to Judge than to Find: Predicting In-Context Learning Success for Demonstration Selection
  36. ICML 2026Efficient Distributed MLLM Training with Cornstarch
  37. ICML 2026Efficient Inference for Noisy LLM-as-a-Judge Evaluation
  38. ICML 2026Enhancing LLM Training via Spectral Clipping
  39. ICML 2026Evaluating Language Models in Realistic Conversational Contexts
  40. ICML 2026ExCyTIn-Bench: Evaluating LLM agents on Cyber Threat Investigation
  41. ICML 2026FairJudge : An Adaptive, Debiased, and Consistent LLM-as-a-Judge
  42. ICML 2026FiRE: Fine-grained Ranking Evaluation for Machine Translation
  43. ICML 2026Game-Theoretic Co-Evolution for LLM-Based Heuristic Discovery
  44. ICML 2026How to Correctly Report LLM-as-a-Judge Evaluations
  45. ICML 2026InnoEval: On Research Idea Evaluation as a Knowledge-Grounded, Multi-Perspective Reasoning Problem
  46. ICML 2026JADE: Expert-Grounded Dynamic Evaluation for Open-Ended Professional Tasks
  47. ICML 2026Judging What We Cannot Solve: A Consequence-Based Approach for Oracle-Free Evaluation of Research-Level Math
  48. ICML 2026Learn-to-learn on Arbitrary Textual Conditioning: A Hypernetwork-Driven Meta-gated LLM
  49. ICML 2026LitReview Arena: Evaluating Literature Review Agents with Battle-style Peer Review Platform
  50. ICML 2026LLM Priors for ERM over Programs

在 Claude、Cursor 里直接问这个趋势

接上 Paperlist MCP 后，写综述、挑方向或准备投稿时，不用离开对话框就能拿到这页的数字：每年各会议的论文数、增长倍数、涨得最快的子方向和代表论文，AI 会按你的问题整理。

可以直接复制这句话问 AI

用 paperlist 查模型即裁判在 NeurIPS 2023–2026 的论文趋势，哪一年涨得最快？

AI 会调用的工具：get_topic_stats {"topic": "llm-as-judge"}

两分钟接入 Paperlist MCP（Claude Code、Cursor、Codex、ChatGPT）

常见问题

2023–2026 年四大会议一共有多少篇模型即裁判论文？
共 401 篇，按年份分别是 2023 年 8 篇、2024 年 48 篇、2025 年 126 篇、2026 年 219 篇。
NeurIPS 2026 有多少篇模型即裁判论文？
NeurIPS 2026 有 88 篇模型即裁判论文，上一届是 53 篇。
ICLR 2026 有多少篇模型即裁判论文？
ICLR 2026 有 62 篇模型即裁判论文，上一届是 37 篇。
ICML 2026 有多少篇模型即裁判论文？
ICML 2026 有 64 篇模型即裁判论文，上一届是 33 篇。
CVPR 2026 有多少篇模型即裁判论文？
CVPR 2026 有 5 篇模型即裁判论文，上一届是 3 篇。
哪个会议的模型即裁判论文最多？
NeurIPS，2023–2026 年共 173 篇。

「评测与基准」下的其他主题