Skip to content
OpenTrain AIFor AI Companies
← Back to explorer

HFEPX Metric Hub

Calibration & Uncertainty Metric Papers In CS.LG

Updated from current HFEPX corpus (Jun 30, 2026). 37 papers are grouped in this metric page.

Read Full Context

Updated from current HFEPX corpus (Jun 30, 2026). 37 papers are grouped in this metric page. Common evaluation modes: Automatic Metrics, Llm As Judge. Most common rater population: Domain Experts. Common annotation unit: Scalar. Frequent quality control: Calibration. Frequently cited benchmark: Deepeval. Common metric signal: auroc. 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 Jun 28, 2026.

Papers: 37 Last published: Jun 28, 2026 Global RSS

When This Metric Page Is Useful

Useful for background comparison, but still validate benchmark and protocol details in the linked papers. Quality band: Medium .

Metric Coverage

13.5%

5 sampled papers include metric names.

Benchmark Anchoring

8.1%

Papers with explicit dataset/benchmark anchors for fair comparison.

Quality Controls

5.4%

2 papers report calibration/adjudication/IAA controls.

  • 37 papers are not low-signal flagged in this sample.
  • Use the protocol matrix below to avoid comparing metrics across incompatible eval setups.

Recommended next step: Treat this as directional signal only; metric reporting is present but benchmark anchoring is still thin.

Main limitation: Benchmark coverage is still thin, so avoid treating this page as a definitive guide to the metric.

What This Metric Page Tells You

What This Metric Page Tells You

  • 8.1% of papers report explicit human-feedback signals, led by expert verification.
  • automatic metrics appears in 13.5% of papers in this hub.
  • Deepeval is a recurring benchmark anchor for cross-paper comparisons in this page.
Metric Notes (Expanded)

Metric-Driven Protocol Takeaways

  • Most common quality-control signal is rater calibration (5.4% of papers).
  • Rater context is mostly domain experts, and annotation is commonly scalar scoring; use this to scope replication staffing.
  • Pair this hub with a human_eval-heavy hub to validate judge-model calibration.

Metric Interpretation

  • auroc is reported in 64.9% of hub papers (24/37); compare with a secondary metric before ranking methods.
  • accuracy is reported in 45.9% of hub papers (17/37); compare with a secondary metric before ranking methods.

Benchmark Context

  • Deepeval appears in 2.7% of hub papers (1/37); use this cohort for benchmark-matched comparisons.
  • DROP appears in 2.7% of hub papers (1/37); use this cohort for benchmark-matched comparisons.

Start Here (Metric-Reliable First 6)

Ranked for metric reporting completeness and comparability.

Metric Protocol Matrix (Top 10)

Compare metric, benchmark, and evaluation context side by side.

Paper Metrics Benchmarks Eval Modes Quality Controls
Deterministic Decisions for High-Stakes AI. A Zero-Egress Pipeline with the Deployability of RAG and the Accuracy of Machine Learning

Jun 28, 2026

 Accuracy, F1 Deepeval Llm As Judge, Automatic Metrics Calibration
Entropy trajectory shape predicts LLM reasoning reliability: A diagnostic study of uncertainty dynamics in chain-of-thought

Mar 19, 2026

 Accuracy, Calibration error GSM8K Automatic Metrics Calibration
FairMed-XGB: A Bayesian-Optimised Multi-Metric Framework with Explainability for Demographic Equity in Critical Healthcare Data

Mar 16, 2026

 Accuracy, Auroc DROP Automatic Metrics Not reported
Do Thinking Tokens Help with Safety?

Jun 23, 2026

 Accuracy, Auroc Not reported Automatic Metrics Not reported
RuleForge: Automated Generation and Validation for Web Vulnerability Detection at Scale

Apr 2, 2026

 Auroc Not reported Llm As Judge, Automatic Metrics Not reported
Just how sure are you? Improving Verbalized Uncertainty Calibration in Medical VQA

Jun 25, 2026

 Not reported Not reported Not reported Not reported
Thermodynamic Signatures of Reasoning: Free-Energy and Spectral-Form-Factor Diagnostics for Hallucination Detection in Large Language Models

Jun 17, 2026

 Not reported Not reported Not reported Not reported
The ACUTE Protocol: Operationalizing Language Model Activations for Better Calibration, Utility, and Trust

Jun 5, 2026

 Not reported Not reported Not reported Not reported
Cognitive Fatigue in Autoregressive Transformers: Formalization and Measurement

May 29, 2026

 Not reported Not reported Not reported Not reported
Amplifying, Not Learning: Fine-Tuned AI Text Detectors Amplify a Pretrained Direction

May 20, 2026

 Not reported Not reported Not reported Not reported
How To Use This Page

Checklist

  • Gap: Papers with explicit human feedback

    Coverage is a replication risk (8.1% vs 45% target).

  • Gap: Papers reporting quality controls

    Coverage is a replication risk (5.4% vs 30% target).

  • Gap: Papers naming benchmarks/datasets

    Coverage is a replication risk (16.2% vs 35% target).

  • Strong: Papers naming evaluation metrics

    Coverage is strong (100% vs 35% target).

  • Gap: Papers with known rater population

    Coverage is a replication risk (5.4% vs 35% target).

  • Gap: Papers with known annotation unit

    Coverage is a replication risk (5.4% vs 35% target).

Strengths

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

Known Gaps

  • Only 5.4% of papers report quality controls; prioritize calibration/adjudication evidence.
  • Rater population is under-specified (5.4% coverage).
  • Annotation unit is under-specified (5.4% coverage).

Suggested Next Analyses

  • Pair this hub with a human_eval-heavy hub to validate judge-model calibration.
  • Stratify by benchmark (Deepeval vs DROP) before comparing methods.
  • Track metric sensitivity by reporting both auroc and accuracy.
  • Add inter-annotator agreement checks when reproducing these protocols.

Recommended Queries

LLM-as-Judge ProtocolsBenchmark Slice: DeepevalMetric Slice: aurocRecent High-Signal Papers
Known Limitations
  • Only 5.4% of papers report quality controls; prioritize calibration/adjudication evidence.
  • Rater population is under-specified (5.4% coverage).
  • Narrative synthesis is grounded in metadata and abstracts only; full-paper implementation details are not parsed.
Coverage Snapshot

Top Metrics

  • Auroc (24)
  • Accuracy (17)
  • Calibration error (14)
  • F1 (4)

Evaluation Modes

  • Automatic Metrics (5)
  • Llm As Judge (2)

Top Benchmarks

  • Deepeval (1)
  • DROP (1)
  • FEVER (1)
  • GSM8K (1)

Agentic Mix

  • Long Horizon (2)

Top Papers Reporting This Metric

Related Metrics And Hubs

Calibration & Uncertainty Metric Papers Auroc In CS.LG Papers CS.LG Human Feedback And Eval Papers Calibration & Uncertainty Metric Papers In CS.CL Calibration & Uncertainty Metric Papers In CS.AI Auroc Metric Papers Coherence + General Metric Papers (16) Helpfulness In CS.CL Papers (17) Helpfulness Metric Papers (20) Relevance + Automatic Metrics Metric Papers (22) Relevance + General Metric Papers (12) Relevance In CS.CL Papers (68) Relevance In CS.LG Papers (21) Relevance Metric Papers (96) Throughput In CS.LG Papers (26) Inference Cost In CS.AI Papers (18)