Skip to content
← Back to explorer

HFEPX Hub

Automatic Metrics + Pairwise Preference Papers

Updated from current HFEPX corpus (Apr 12, 2026). 101 papers are grouped in this hub page.

Read Full Context

Updated from current HFEPX corpus (Apr 12, 2026). 101 papers are grouped in this hub page. Common evaluation modes: Automatic Metrics, Llm As Judge. Most common rater population: Domain Experts. Common annotation unit: Pairwise. Frequent quality control: Inter Annotator Agreement Reported. Frequently cited benchmark: AlpacaEval. 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 Jul 15, 2025.

Papers: 101 Last published: Jul 15, 2025 Global RSS Tag RSS
Automatic MetricsPairwise Preference

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 101 total papers in this hub).

High-Signal Coverage

100.0%

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

Replication-Ready Set

20

Benchmark + metric + eval mode explicitly present.

Judge/Human Comparability

0

Papers containing both `human_eval` and `llm_as_judge`.

  • 20 papers are replication-ready (benchmark + metric + explicit evaluation mode).
  • 0 papers support judge-vs-human agreement analysis.
  • 7 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.

Need evaluators for this research workflow?

Post a Job →

Why This Matters For Eval Research

  • 100% of papers report explicit human-feedback signals, led by pairwise preferences.
  • automatic metrics appears in 100% of papers in this hub.
  • AlpacaEval is a recurring benchmark anchor for cross-paper comparisons in this page.

Protocol Takeaways

  • Most common quality-control signal is inter-annotator agreement reporting (5% of papers).
  • Rater context is mostly domain experts, and annotation is commonly pairwise annotation; use this to scope replication staffing.
  • Compare papers that report both human_eval and llm_as_judge to quantify judge-human agreement drift.

Benchmark Interpretation

  • AlpacaEval appears in 3% of hub papers (3/101); use this cohort for benchmark-matched comparisons.
  • GSM8K appears in 3% of hub papers (3/101); use this cohort for benchmark-matched comparisons.

Metric Interpretation

  • accuracy is reported in 45.5% of hub papers (46/101); compare with a secondary metric before ranking methods.
  • cost is reported in 14.9% of hub papers (15/101); compare with a secondary metric before ranking methods.
Researcher Checklist (Expanded)

Researcher Checklist

  • Strong: Papers with explicit human feedback

    Coverage is strong (100% vs 45% target).

  • Gap: Papers reporting quality controls

    Coverage is a replication risk (6.9% vs 30% target).

  • Gap: Papers naming benchmarks/datasets

    Coverage is a replication risk (20.8% vs 35% target).

  • Strong: Papers naming evaluation metrics

    Coverage is strong (90.1% vs 35% target).

  • Gap: Papers with known rater population

    Coverage is a replication risk (10.9% vs 35% target).

  • Strong: Papers with known annotation unit

    Coverage is strong (48.5% vs 35% target).

Strengths

  • Strong human-feedback signal (100% of papers).
  • Contains both human-eval and LLM-as-judge protocols for head-to-head methodology comparison.

Known Gaps

  • Only 6.9% of papers report quality controls; prioritize calibration/adjudication evidence.
  • Rater population is under-specified (10.9% coverage).

Suggested Next Analyses

  • Compare papers that report both human_eval and llm_as_judge to quantify judge-human agreement drift.
  • Stratify by benchmark (AlpacaEval vs GSM8K) before comparing methods.
  • Track metric sensitivity by reporting both accuracy and cost.
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
SCOPE: Selective Conformal Optimized Pairwise LLM Judging

Feb 13, 2026

Yes Automatic Metrics MT Bench , LMSYS Chatbot Arena Error rate Calibration
Personalized RewardBench: Evaluating Reward Models with Human Aligned Personalization

Apr 8, 2026

Yes Human Eval , Automatic Metrics Rewardbench Accuracy , Helpfulness Not Reported
Modeling and Benchmarking Spoken Dialogue Rewards with Modality and Colloquialness

Mar 16, 2026

Yes Automatic Metrics Esdr Bench Accuracy Not Reported
Do Phone-Use Agents Respect Your Privacy?

Apr 1, 2026

Yes Automatic Metrics APPS , Myphonebench Task success Not Reported
ClimateCheck 2026: Scientific Fact-Checking and Disinformation Narrative Classification of Climate-related Claims

Mar 27, 2026

Yes Automatic Metrics Codabench Recall , Recall@k Not Reported
Stabilizing Iterative Self-Training with Verified Reasoning via Symbolic Recursive Self-Alignment

Mar 23, 2026

Yes Automatic Metrics GSM8K Accuracy Not Reported
DSPA: Dynamic SAE Steering for Data-Efficient Preference Alignment

Mar 23, 2026

Yes Automatic Metrics MT Bench , AlpacaEval Accuracy Not Reported
CausalRM: Causal-Theoretic Reward Modeling for RLHF from Observational User Feedbacks

Mar 19, 2026

Yes Automatic Metrics Harmbench Cost Not Reported
Elo-Evolve: A Co-evolutionary Framework for Language Model Alignment

Feb 14, 2026

Yes Automatic Metrics MT Bench , AlpacaEval Elo Not Reported
PEARL: Self-Evolving Assistant for Time Management with Reinforcement Learning

Jan 17, 2026

Yes Automatic Metrics Calconflictbench Error rate Not Reported
MemoryArena: Benchmarking Agent Memory in Interdependent Multi-Session Agentic Tasks

Feb 18, 2026

Yes Automatic Metrics Memoryarena Recall Not Reported
From Consensus to Split Decisions: ABC-Stratified Sentiment in Holocaust Oral Histories

Mar 30, 2026

Yes Automatic Metrics Not Reported Kappa , Agreement Inter Annotator Agreement Reported

Protocol Diff (Top Papers)

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

Signal SCOPE: Selective Conformal Optimized Pairwise LLM J… Personalized RewardBench: Evaluating Reward Models… Modeling and Benchmarking Spoken Dialogue Rewards w…
Human Feedback Pairwise PreferencePairwise Preference, Rubric RatingPairwise Preference
Evaluation Modes Automatic MetricsHuman Eval, Automatic MetricsAutomatic Metrics
Benchmarks MT Bench, LMSYS Chatbot ArenaRewardbenchEsdr Bench
Metrics Error rateAccuracy, HelpfulnessAccuracy
Quality Controls CalibrationNot reportedNot reported
Rater Population UnknownUnknownUnknown
Annotation Unit PairwisePairwisePairwise
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. Personalized RewardBench: Evaluating Reward Models with Human Aligned Personalization

    Start here for detailed protocol reporting and quality-control evidence. Signals: human evaluation + pairwise preferences. Focus: Rewardbench / accuracy. Abstract: While benchmarks for general response quality are prevalent,.

  3. MMEmb-R1: Reasoning-Enhanced Multimodal Embedding with Pair-Aware Selection and Adaptive Control

    Start here for detailed protocol reporting and quality-control evidence. Signals: automatic metrics + pairwise preferences. Focus: latency. Abstract: First, structural misalignment between instance-level reasoning and pairwise contrastive supervision.

  4. Let's Think in Two Steps: Mitigating Agreement Bias in MLLMs with Self-Grounded Verification

    Include a human-eval paper to calibrate against judge-based evaluation settings. Signals: automatic metrics + pairwise preferences. Focus: VisualWebArena / accuracy. Abstract: Multimodal LLMs (MLLMs) offer a promising solution,.

  5. Blinded Radiologist and LLM-Based Evaluation of LLM-Generated Japanese Translations of Chest CT Reports: Comparative Study

    Include an LLM-as-judge paper to test judge design and agreement assumptions. Signals: LLM-as-judge + pairwise preferences. Focus: accuracy. Abstract: Objective: To evaluate the educational suitability of LLM-generated Japanese.

  6. SCOPE: Selective Conformal Optimized Pairwise LLM Judging

    Adds automatic metrics with pairwise preferences for broader protocol coverage within this hub. Signals: automatic metrics + pairwise preferences. Focus: MT-Bench / error rate. Abstract: Large language models.

  7. Elo-Evolve: A Co-evolutionary Framework for Language Model Alignment

    Adds automatic metrics with pairwise preferences for broader protocol coverage within this hub. Signals: automatic metrics + pairwise preferences. Focus: MT-Bench / elo. Abstract: Current alignment methods for.

  8. PEARL: Self-Evolving Assistant for Time Management with Reinforcement Learning

    Adds automatic metrics with pairwise preferences for broader protocol coverage within this hub. Signals: automatic metrics + pairwise preferences. Focus: Calconflictbench / error rate. Abstract: We refer to.

Known Limitations

Known Limitations

  • Only 6.9% of papers report quality controls; prioritize calibration/adjudication evidence.
  • Rater population is under-specified (10.9% coverage).
  • Narrative synthesis is grounded in metadata and abstracts only; full-paper implementation details are not parsed.
Research Utility Snapshot (Detailed)

Research Utility Snapshot

Human Feedback Mix

  • Pairwise Preference (101)
  • Rubric Rating (4)
  • Critique Edit (3)
  • Expert Verification (2)

Evaluation Modes

  • Automatic Metrics (101)
  • Llm As Judge (2)
  • Human Eval (1)
  • Simulation Env (1)

Top Benchmarks

  • AlpacaEval (3)
  • GSM8K (3)
  • MT Bench (3)
  • AIME (2)

Top Metrics

  • Accuracy (46)
  • Cost (15)
  • Relevance (10)
  • Agreement (9)

Rater Population Mix

  • Domain Experts (11)

Quality Controls

  • Inter Annotator Agreement Reported (5)
  • Calibration (2)
  • Adjudication (1)
Coverage diagnostics (sample-based): human-feedback 100.0% · benchmarks 33.3% · metrics 100.0% · quality controls 11.7%.

Top Papers

Related Hubs

Need human evaluators for your AI research? Scale annotation with expert AI Trainers.