Skip to content
← Back to explorer

HFEPX Archive Slice

HFEPX Weekly Archive: 2026-W13

Updated from current HFEPX corpus (Apr 12, 2026). 677 papers are grouped in this daily page.

Read Full Context

Updated from current HFEPX corpus (Apr 12, 2026). 677 papers are grouped in this daily page. Common evaluation modes: Automatic Metrics, Simulation Env. Most common rater population: Domain Experts. Common annotation unit: Trajectory. Frequent quality control: Calibration. Frequently cited benchmark: GSM8K. 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 29, 2026.

Papers: 677 Last published: Mar 29, 2026 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 677 papers).

High-Signal Coverage

100.0%

60 / 60 papers are not low-signal flagged.

Benchmark Anchors

13.3%

Papers with benchmark/dataset mentions in extraction output.

Metric Anchors

38.3%

Papers with reported metric mentions in extraction output.

  • 2 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

  • 6.1% of papers report explicit human-feedback signals, led by pairwise preferences.
  • automatic metrics appears in 20.1% of papers in this hub.
  • GSM8K 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 (1% of papers).
  • Rater context is mostly domain experts, and annotation is commonly trajectory-level annotation; use this to scope replication staffing.
  • Compare papers that report both human_eval and llm_as_judge to quantify judge-human agreement drift.

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

Mar 28, 2026

Llm As Judge, Automatic Metrics MMLU Accuracy, Relevance Not reported
Retromorphic Testing with Hierarchical Verification for Hallucination Detection in RAG

Mar 29, 2026

Automatic Metrics RAGTruth F1, Faithfulness Not reported
The Geometry of Harmful Intent: Training-Free Anomaly Detection via Angular Deviation in LLM Residual Streams

Mar 28, 2026

Automatic Metrics XSTest Nll, Auroc Not reported
LLM Readiness Harness: Evaluation, Observability, and CI Gates for LLM/RAG Applications

Mar 28, 2026

Automatic Metrics BEIR Latency, Latency p95 Not reported
Improving Clinical Diagnosis with Counterfactual Multi-Agent Reasoning

Mar 29, 2026

Human Eval, Automatic Metrics Not reported Accuracy Not reported
PRBench: End-to-end Paper Reproduction in Physics Research

Mar 29, 2026

Automatic Metrics, Simulation Env Not reported Accuracy, Success rate Not reported
Routing Sensitivity Without Controllability: A Diagnostic Study of Fairness in MoE Language Models

Mar 28, 2026

Automatic Metrics Not reported Cost Not reported
KazByte: Adapting Qwen models to Kazakh via Byte-level Adapter

Mar 29, 2026

Automatic Metrics Not reported Accuracy Not reported
Model Capability Dominates: Inference-Time Optimization Lessons from AIMO 3

Mar 29, 2026

Automatic Metrics Not reported Accuracy Not reported
KVSculpt: KV Cache Compression as Distillation

Mar 29, 2026

Automatic Metrics Not reported Mse, Cost Not reported
Researcher Workflow (Detailed)

Checklist

  • Gap: Papers with explicit human feedback

    Coverage is a replication risk (6.1% vs 45% target).

  • Gap: Papers reporting quality controls

    Coverage is a replication risk (1.9% vs 30% target).

  • Gap: Papers naming benchmarks/datasets

    Coverage is a replication risk (8.1% vs 35% target).

  • Moderate: Papers naming evaluation metrics

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

  • Gap: Papers with known rater population

    Coverage is a replication risk (6.8% vs 35% target).

  • Gap: Papers with known annotation unit

    Coverage is a replication risk (6.2% vs 35% target).

Strengths

  • Contains both human-eval and LLM-as-judge protocols for head-to-head methodology comparison.

Known Gaps

  • Only 1.9% of papers report quality controls; prioritize calibration/adjudication evidence.
  • Rater population is under-specified (6.8% coverage).
  • Annotation unit is under-specified (6.2% coverage).

Suggested Next Analyses

  • Compare papers that report both human_eval and llm_as_judge to quantify judge-human agreement drift.
  • Stratify by benchmark (GSM8K vs DROP) before comparing methods.
  • Track metric sensitivity by reporting both accuracy and cost.

Recommended Queries

Known Limitations
  • Only 1.9% of papers report quality controls; prioritize calibration/adjudication evidence.
  • Rater population is under-specified (6.8% 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 (136)
  • Simulation Env (13)
  • Human Eval (9)
  • Llm As Judge (9)

Top Metrics

  • Accuracy (86)
  • Cost (41)
  • Precision (19)
  • F1 (17)

Top Benchmarks

  • GSM8K (5)
  • DROP (4)
  • MMLU (4)
  • BIRD (2)

Quality Controls

  • Calibration (7)
  • Adjudication (2)
  • Gold Questions (2)
  • Inter Annotator Agreement Reported (2)

Papers In This Archive Slice

Recent Archive Slices

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