Skip to content
← Back to explorer

HFEPX Archive Slice

HFEPX Weekly Archive: 2025-W39

Updated from current HFEPX corpus (Apr 12, 2026). 90 papers are grouped in this daily page.

Read Full Context

Updated from current HFEPX corpus (Apr 12, 2026). 90 papers are grouped in this daily page. Common evaluation modes: Automatic Metrics, Llm As Judge. Most common rater population: Domain Experts. Common annotation unit: Trajectory. Frequent quality control: Calibration. Frequently cited benchmark: MT-Bench. 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 Sep 28, 2025.

Papers: 90 Last published: Sep 28, 2025 Global RSS

Researcher Quick Triage

Use this archive page for time-slice monitoring (what changed in evaluation methods, metrics, and protocol quality this period). Quality band: High .

Analysis blocks are computed from the loaded sample (60 of 90 papers).

High-Signal Coverage

100.0%

60 / 60 papers are not low-signal flagged.

Benchmark Anchors

10.0%

Papers with benchmark/dataset mentions in extraction output.

Metric Anchors

41.7%

Papers with reported metric mentions in extraction output.

  • 4 papers report explicit quality controls for this archive period.
  • Prioritize papers with both benchmark and metric anchors for reliable longitudinal comparisons.

Primary action: Use this slice for trend comparison: review top papers first, then validate shifts in the protocol matrix.

Get this digest every Friday →

Subscribe

Why This Time Slice Matters

  • 14.4% of papers report explicit human-feedback signals, led by pairwise preferences.
  • automatic metrics appears in 34.4% of papers in this hub.
  • MT-Bench is a recurring benchmark anchor for cross-paper comparisons in this page.

Protocol Takeaways For This Period

  • Most common quality-control signal is rater calibration (3.3% of papers).
  • Rater context is mostly domain experts, and annotation is commonly trajectory-level annotation; use this to scope replication staffing.
  • Pair this hub with a human_eval-heavy hub to validate judge-model calibration.

Start Here (Highest-Signal Papers In This Slice)

Ranked by protocol completeness and evidence density for faster period-over-period review.

Protocol Matrix (Top 10)

Quickly compare method ingredients across this archive slice.

Paper Eval Modes Benchmarks Metrics Quality Controls
IA2: Alignment with ICL Activations Improves Supervised Fine-Tuning

Sep 26, 2025

Automatic Metrics Not reported Accuracy, Cost Calibration
Induction Signatures Are Not Enough: A Matched-Compute Study of Load-Bearing Structure in In-Context Learning

Sep 26, 2025

Automatic Metrics DROP Perplexity Not reported
HEART: Emotionally-Driven Test-Time Scaling of Language Models

Sep 26, 2025

Automatic Metrics GPQA, LiveCodeBench Accuracy Not reported
LogiPart: Local Large Language Models for Data Exploration at Scale with Logical Partitioning

Sep 26, 2025

Llm As Judge, Automatic Metrics Not reported Accuracy Calibration
CoSpaDi: Compressing LLMs via Calibration-Guided Sparse Dictionary Learning

Sep 26, 2025

Automatic Metrics Not reported Accuracy, Perplexity Calibration
ReviewScore: Misinformed Peer Review Detection with Large Language Models

Sep 25, 2025

Automatic Metrics Not reported F1, Kappa Inter Annotator Agreement Reported
Uncovering Grounding IDs: How External Cues Shape Multimodal Binding

Sep 28, 2025

Automatic Metrics Not reported Accuracy Not reported
SPELL: Self-Play Reinforcement Learning for Evolving Long-Context Language Models

Sep 28, 2025

Automatic Metrics Not reported Pass@8 Not reported
From What to Why: A Multi-Agent System for Evidence-based Chemical Reaction Condition Reasoning

Sep 28, 2025

Automatic Metrics Not reported Accuracy, Recall Not reported
Mapping Overlaps in Benchmarks through Perplexity in the Wild

Sep 27, 2025

Automatic Metrics Not reported Perplexity Not reported
Researcher Workflow (Detailed)

Checklist

  • Gap: Papers with explicit human feedback

    Coverage is a replication risk (14.4% vs 45% target).

  • Gap: Papers reporting quality controls

    Coverage is a replication risk (4.4% vs 30% target).

  • Gap: Papers naming benchmarks/datasets

    Coverage is a replication risk (5.6% vs 35% target).

  • Gap: Papers naming evaluation metrics

    Coverage is a replication risk (20% vs 35% target).

  • Gap: Papers with known rater population

    Coverage is a replication risk (7.8% vs 35% target).

  • Gap: Papers with known annotation unit

    Coverage is a replication risk (4.4% vs 35% target).

Strengths

  • This hub still surfaces a concentrated paper set for protocol triage and replication planning.

Known Gaps

  • Only 4.4% of papers report quality controls; prioritize calibration/adjudication evidence.
  • Rater population is under-specified (7.8% coverage).
  • Annotation unit is under-specified (4.4% coverage).

Suggested Next Analyses

  • Pair this hub with a human_eval-heavy hub to validate judge-model calibration.
  • Stratify by benchmark (MT-Bench vs AlpacaEval) before comparing methods.
  • Track metric sensitivity by reporting both accuracy and cost.

Recommended Queries

Known Limitations
  • Only 4.4% of papers report quality controls; prioritize calibration/adjudication evidence.
  • Rater population is under-specified (7.8% coverage).
  • Narrative synthesis is grounded in metadata and abstracts only; full-paper implementation details are not parsed.
Research Utility Snapshot (Detailed)

Evaluation Modes

  • Automatic Metrics (31)
  • Llm As Judge (3)
  • Simulation Env (3)

Top Metrics

  • Accuracy (12)
  • Cost (3)
  • Agreement (1)
  • Error rate (1)

Top Benchmarks

  • MT Bench (2)
  • AlpacaEval (1)
  • AlpacaEval 2.0 (1)
  • Arena Hard (1)

Quality Controls

  • Calibration (3)
  • Inter Annotator Agreement Reported (1)

Papers In This Archive Slice

Recent Archive Slices

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.