Skip to content
← Back to explorer

HFEPX Hub

CS.CL + Llm As Judge Papers

Updated from current HFEPX corpus (Apr 12, 2026). 103 papers are grouped in this hub page.

Read Full Context

Updated from current HFEPX corpus (Apr 12, 2026). 103 papers are grouped in this hub page. Common evaluation modes: Llm As Judge, Automatic Metrics. Most common rater population: Domain Experts. Common annotation unit: Multi Dim Rubric. Frequent quality control: Calibration. Frequently cited benchmark: Healthbench. Common metric signal: accuracy. Use this page to compare protocol setup, judge behavior, and labeling design decisions before running new eval experiments. Newest paper in this set is from Mar 22, 2026.

Papers: 103 Last published: Mar 22, 2026 Global RSS Tag RSS
Cs.CLLlm As Judge

Researcher Quick Triage

This hub is best used for protocol triage and replication planning from abstract-level evidence. Quality band: High .

Analysis blocks below are computed from the currently loaded sample (60 of 103 total papers in this hub).

High-Signal Coverage

100.0%

60 / 60 sampled papers are not low-signal flagged.

Replication-Ready Set

13

Benchmark + metric + eval mode explicitly present.

Judge/Human Comparability

4

Papers containing both `human_eval` and `llm_as_judge`.

  • 13 papers are replication-ready (benchmark + metric + explicit evaluation mode).
  • 4 papers support judge-vs-human agreement analysis.
  • 6 papers report explicit quality controls (calibration/adjudication/IAA).

Primary action: Start with the top 2 papers in “Start Here”, then validate assumptions in the protocol matrix.

Currently showing only replication-ready papers in ranking and matrix sections (13 papers).

Need evaluators for this research workflow?

Post a Job →

Why This Matters For Eval Research

  • 48.7% of papers report explicit human-feedback signals, led by pairwise preferences.
  • LLM-as-judge appears in 75.7% of papers in this hub.
  • Healthbench is a recurring benchmark anchor for cross-paper comparisons in this page.

Protocol Takeaways

  • 2 sampled papers report both human evaluation and LLM-as-judge, supporting direct agreement checks.
  • Most common quality-control signal is rater calibration (2.9% of papers).
  • Rater context is mostly domain experts, and annotation is commonly multi-dimensional rubrics; use this to scope replication staffing.

Benchmark Interpretation

  • Healthbench appears in 2.6% of hub papers (2/103); use this cohort for benchmark-matched comparisons.
  • AdvBench appears in 1.3% of hub papers (1/103); use this cohort for benchmark-matched comparisons.

Metric Interpretation

  • accuracy is reported in 28.2% of hub papers (22/103); compare with a secondary metric before ranking methods.
  • agreement is reported in 10.3% of hub papers (8/103); compare with a secondary metric before ranking methods.
Researcher Checklist (Expanded)

Researcher Checklist

  • Strong: Papers with explicit human feedback

    Coverage is strong (48.7% vs 45% target).

  • Gap: Papers reporting quality controls

    Coverage is a replication risk (7.7% vs 30% target).

  • Moderate: Papers naming benchmarks/datasets

    Coverage is usable but incomplete (30.8% vs 35% target).

  • Strong: Papers naming evaluation metrics

    Coverage is strong (56.4% vs 35% target).

  • Moderate: Papers with known rater population

    Coverage is usable but incomplete (28.2% vs 35% target).

  • Moderate: Papers with known annotation unit

    Coverage is usable but incomplete (29.5% vs 35% target).

Strengths

  • Strong human-feedback signal (48.7% of papers).
  • Contains both human-eval and LLM-as-judge protocols for head-to-head methodology comparison.

Known Gaps

  • Only 7.7% of papers report quality controls; prioritize calibration/adjudication evidence.

Suggested Next Analyses

  • Compare papers that report both human_eval and llm_as_judge to quantify judge-human agreement drift.
  • Stratify by benchmark (Healthbench vs AdvBench) before comparing methods.
  • Track metric sensitivity by reporting both accuracy and agreement.
Recommended Queries (Expanded)

Recommended Queries

Start with These 3

Use these when you need one protocol anchor, one benchmark anchor, and one recent comparison point before reading the wider hub.

Start Here (Best First 6)

Ranked for protocol completeness (human signal, benchmark + metric anchors, quality controls, and judge/human overlap).

Protocol Matrix (Top 12)

Use this to quickly compare protocol ingredients instead of scanning long prose.

Paper HF Signal Eval Modes Benchmarks Metrics QC
AgentHER: Hindsight Experience Replay for LLM Agent Trajectory Relabeling

Mar 22, 2026

Yes Human Eval , Llm As Judge WebArena , ToolBench Precision , Pass@1 Not Reported
PoSh: Using Scene Graphs To Guide LLMs-as-a-Judge For Detailed Image Descriptions

Oct 21, 2025

Yes Human Eval , Llm As Judge CAPArena Spearman Not Reported
PubMed Reasoner: Dynamic Reasoning-based Retrieval for Evidence-Grounded Biomedical Question Answering

Mar 28, 2026

Yes Llm As Judge , Automatic Metrics MMLU Accuracy , Relevance Not Reported
PanCanBench: A Comprehensive Benchmark for Evaluating Large Language Models in Pancreatic Oncology

Mar 2, 2026

Yes Llm As Judge , Automatic Metrics Pancanbench , Healthbench Accuracy Not Reported
Jailbreak Foundry: From Papers to Runnable Attacks for Reproducible Benchmarking

Feb 27, 2026

Yes Llm As Judge AdvBench , Jbf Eval Success rate , Jailbreak success rate Not Reported
LLM-as-a-Judge for Time Series Explanations

Apr 2, 2026

No
Not Reported
Llm As Judge , Automatic Metrics DROP Accuracy , Faithfulness Not Reported
Weakly Supervised Distillation of Hallucination Signals into Transformer Representations

Apr 7, 2026

No
Not Reported
Llm As Judge , Automatic Metrics SQuAD F1 , Latency Not Reported
No Free Labels: Limitations of LLM-as-a-Judge Without Human Grounding

Mar 7, 2025

Yes Llm As Judge MT Bench , Bff Bench Agreement , Cost Not Reported
LIT-RAGBench: Benchmarking Generator Capabilities of Large Language Models in Retrieval-Augmented Generation

Mar 6, 2026

No
Not Reported
Llm As Judge , Automatic Metrics Lit Ragbench Accuracy Not Reported
Reward Prediction with Factorized World States

Mar 10, 2026

No
Not Reported
Llm As Judge , Simulation Env ALFWorld Success rate Not Reported
ThaiSafetyBench: Assessing Language Model Safety in Thai Cultural Contexts

Mar 5, 2026

No
Not Reported
Llm As Judge , Automatic Metrics Thaisafetybench F1 , F1 weighted Not Reported
LiveMCPBench: Can Agents Navigate an Ocean of MCP Tools?

Aug 3, 2025

No
Not Reported
Llm As Judge Livemcpbench Task success Not Reported

Protocol Diff (Top Papers)

Fast side-by-side comparison for the highest-ranked papers in this hub.

Signal AgentHER: Hindsight Experience Replay for LLM Agent… PoSh: Using Scene Graphs To Guide LLMs-as-a-Judge F… PubMed Reasoner: Dynamic Reasoning-based Retrieval…
Human Feedback DemonstrationsRubric RatingExpert Verification
Evaluation Modes Human Eval, Llm As JudgeHuman Eval, Llm As JudgeLlm As Judge, Automatic Metrics
Benchmarks WebArena, ToolBenchCAPArenaMMLU
Metrics Precision, Pass@1SpearmanAccuracy, Relevance
Quality Controls Not reportedNot reportedNot reported
Rater Population UnknownDomain ExpertsDomain Experts
Annotation Unit TrajectoryMulti Dim RubricUnknown
Suggested Reading Order (Extended)

This section is intentionally expanded only when needed; use “Start Here” above for a faster pass.

Suggested Reading Order

  1. HyperMem: Hypergraph Memory for Long-Term Conversations

    Start here for detailed protocol reporting and quality-control evidence. Signals: LLM-as-judge + pairwise preferences. Focus: accuracy. Abstract: However, existing approaches as Retrieval-Augmented Generation (RAG) and graph-based memory mostly.

  2. Self-Preference Bias in Rubric-Based Evaluation of Large Language Models

    Start here for detailed protocol reporting and quality-control evidence. Signals: LLM-as-judge + pairwise preferences. Focus: IFEval. Abstract: LLM-as-a-judge has become the de facto approach for evaluating LLM outputs.

  3. AgentHER: Hindsight Experience Replay for LLM Agent Trajectory Relabeling

    Include a human-eval paper to calibrate against judge-based evaluation settings. Signals: human evaluation + demonstration data. Focus: WebArena / precision. Abstract: AgentHER realises this idea through a four-stage.

  4. PoSh: Using Scene Graphs To Guide LLMs-as-a-Judge For Detailed Image Descriptions

    Include a human-eval paper to calibrate against judge-based evaluation settings. Signals: human evaluation + rubric ratings. Focus: CAPArena / spearman. Abstract: In this work, we introduce PoSh, a.

  5. Jailbreak Foundry: From Papers to Runnable Attacks for Reproducible Benchmarking

    Include an LLM-as-judge paper to test judge design and agreement assumptions. Signals: LLM-as-judge + red-team protocols. Focus: AdvBench / success rate. Abstract: This system enables a standardized AdvBench.

  6. PanCanBench: A Comprehensive Benchmark for Evaluating Large Language Models in Pancreatic Oncology

    Adds LLM-as-judge with rubric ratings for broader protocol coverage within this hub. Signals: LLM-as-judge + rubric ratings. Focus: Pancanbench / accuracy. Abstract: Moreover, high rubric-based scores do not.

  7. PubMed Reasoner: Dynamic Reasoning-based Retrieval for Evidence-Grounded Biomedical Question Answering

    Adds LLM-as-judge with expert verification for broader protocol coverage within this hub. Signals: LLM-as-judge + expert verification. Focus: MMLU / accuracy. Abstract: Moreover, LLM-as-judge evaluations prefer our responses.

Known Limitations

Known Limitations

  • Only 7.7% of papers report quality controls; prioritize calibration/adjudication evidence.
  • Narrative synthesis is grounded in metadata and abstracts only; full-paper implementation details are not parsed.
  • Cross-page comparisons should be benchmark- and metric-matched to avoid protocol confounding.
Research Utility Snapshot (Detailed)

Research Utility Snapshot

Human Feedback Mix

  • Pairwise Preference (20)
  • Rubric Rating (12)
  • Expert Verification (7)
  • Red Team (4)

Evaluation Modes

  • Llm As Judge (78)
  • Automatic Metrics (33)
  • Human Eval (10)
  • Simulation Env (6)

Top Benchmarks

  • Healthbench (2)
  • AdvBench (1)
  • ALFWorld (1)
  • Aurora Bench (1)

Top Metrics

  • Accuracy (22)
  • Agreement (8)
  • Cost (6)
  • F1 (6)

Rater Population Mix

  • Domain Experts (22)

Quality Controls

  • Calibration (3)
  • Adjudication (2)
  • Inter Annotator Agreement Reported (1)
Coverage diagnostics (sample-based): human-feedback 65.0% · benchmarks 38.3% · metrics 51.7% · quality controls 10.0%.

Top Papers

  • AgentHER: Hindsight Experience Replay for LLM Agent Trajectory Relabeling

    Liang Ding · Mar 22, 2026 · Citations: 0

    Demonstrations Human EvalLlm As Judge Long Horizon

    LLM agents fail on the majority of real-world tasks -- GPT-4o succeeds on fewer than 15% of WebArena navigation tasks and below 55% pass@1 on ToolBench (Zhou et al., 2024; Qin et al., 2024) -- yet every failed trajectory is routinely…

  • PanCanBench: A Comprehensive Benchmark for Evaluating Large Language Models in Pancreatic Oncology

    Yimin Zhao, Sheela R. Damle, Simone E. Dekker, Scott Geng, Karly Williams Silva · Mar 2, 2026 · Citations: 0

    Rubric RatingExpert Verification Llm As JudgeAutomatic Metrics

    Large language models (LLMs) have achieved expert-level performance on standardized examinations, yet multiple-choice accuracy poorly reflects real-world clinical utility and safety.

  • PoSh: Using Scene Graphs To Guide LLMs-as-a-Judge For Detailed Image Descriptions

    Amith Ananthram, Elias Stengel-Eskin, Lorena A. Bradford, Julia Demarest, Adam Purvis · Oct 21, 2025 · Citations: 0

    Rubric Rating Human EvalLlm As Judge

    In this work, we introduce PoSh, a metric for detailed image description that uses scene graphs as structured rubrics to guide LLMs-as-a-Judge, producing aggregate scores grounded in fine-grained errors (e.g.

  • Jailbreak Foundry: From Papers to Runnable Attacks for Reproducible Benchmarking

    Zhicheng Fang, Jingjie Zheng, Chenxu Fu, Wei Xu · Feb 27, 2026 · Citations: 0

    Red Team Llm As Judge Multi Agent

    Jailbreak techniques for large language models (LLMs) evolve faster than benchmarks, making robustness estimates stale and difficult to compare across papers due to drift in datasets, harnesses, and judging protocols.

  • PubMed Reasoner: Dynamic Reasoning-based Retrieval for Evidence-Grounded Biomedical Question Answering

    Yiqing Zhang, Xiaozhong Liu, Fabricio Murai · Mar 28, 2026 · Citations: 0

    Expert Verification Llm As JudgeAutomatic Metrics

    In this context, we introduce PubMed Reasoner, a biomedical QA agent composed of three stages: self-critic query refinement evaluates MeSH terms for coverage, alignment, and redundancy to enhance PubMed queries based on partial (metadata)…

  • No Free Labels: Limitations of LLM-as-a-Judge Without Human Grounding

    Michael Krumdick, Charles Lovering, Varshini Reddy, Seth Ebner, Chris Tanner · Mar 7, 2025 · Citations: 0

    Pairwise Preference Llm As Judge

    To address this gap, we introduce the Business and Finance Fundamentals Benchmark (BFF-Bench), a dataset of 160 challenging questions and long-form responses authored by financial professionals.

  • LIT-RAGBench: Benchmarking Generator Capabilities of Large Language Models in Retrieval-Augmented Generation

    Koki Itai, Shunichi Hasegawa, Yuta Yamamoto, Gouki Minegishi, Masaki Otsuki · Mar 6, 2026 · Citations: 0

    Llm As JudgeAutomatic Metrics Long Horizon

    To bridge the gap between existing evaluations and practical use, we introduce LIT-RAGBench (the Logic, Integration, Table, Reasoning, and Abstention RAG Generator Benchmark), which defines five categories: Integration, Reasoning, Logic,…

  • LiveMCPBench: Can Agents Navigate an Ocean of MCP Tools?

    Guozhao Mo, Wenliang Zhong, Jiawei Chen, Qianhao Yuan, Xuanang Chen · Aug 3, 2025 · Citations: 0

    Llm As Judge Tool Use

    Unfortunately, there is still a large gap between real-world MCP usage and current evaluation: they typically assume single-server settings and directly inject tools into the model's context, bypassing the challenges of large-scale…

  • LLM-as-a-Judge for Time Series Explanations

    Preetham Sivalingam, Murari Mandal, Saurabh Deshpande, Dhruv Kumar · Apr 2, 2026 · Citations: 0

    Llm As JudgeAutomatic Metrics

    Although modern models generate textual interpretations of numerical signals, existing evaluation methods are limited: reference based similarity metrics and consistency checking models require ground truth explanations, while traditional…

  • Weakly Supervised Distillation of Hallucination Signals into Transformer Representations

    Shoaib Sadiq Salehmohamed, Jinal Prashant Thakkar, Hansika Aredla, Shaik Mohammed Omar, Shalmali Ayachit · Apr 7, 2026 · Citations: 0

    Llm As JudgeAutomatic Metrics

    We introduce a weak supervision framework that combines three complementary grounding signals: substring matching, sentence embedding similarity, and an LLM as a judge verdict to label generated responses as grounded or hallucinated without…

  • Reward Prediction with Factorized World States

    Yijun Shen, Delong Chen, Xianming Hu, Jiaming Mi, Hongbo Zhao · Mar 10, 2026 · Citations: 0

    Llm As JudgeSimulation Env

    We evaluate on RewardPrediction, a new benchmark dataset spanning five diverse domains and comprising 2,454 unique action-observation trajectories with step-wise ground-truth rewards.

  • ThaiSafetyBench: Assessing Language Model Safety in Thai Cultural Contexts

    Trapoom Ukarapol, Nut Chukamphaeng, Kunat Pipatanakul, Pakhapoom Sarapat · Mar 5, 2026 · Citations: 0

    Llm As JudgeAutomatic Metrics

    Using ThaiSafetyBench, we evaluate 24 LLMs, with GPT-4.1 and Gemini-2.5-Pro serving as LLM-as-a-judge evaluators.

  • Agri-Query: A Case Study on RAG vs. Long-Context LLMs for Cross-Lingual Technical Question Answering

    Julius Gun, Timo Oksanen · Aug 25, 2025 · Citations: 0

    Llm As JudgeAutomatic Metrics

    Our benchmark is built on a user manual for an agricultural machine, available in English, French, and German.

Related Hubs

Get Started

Join the #1 Platform for AI Training Talent

Where top AI builders and expert AI Trainers connect to build the future of AI.
Self-Service
Post a Job
Post your project and get a shortlist of qualified AI Trainers and Data Labelers. Hire and manage your team in the tools you already use.
Managed Service
For Large Projects
Done-for-You
We recruit, onboard, and manage a dedicated team inside your tools. End-to-end operations for large or complex projects.
For Freelancers
Join as an AI Trainer
Find AI training and data labeling projects across platforms, all in one place. One profile, one application process, more opportunities.