Skip to content
← Back to explorer

HFEPX Hub

CS.LG + Automatic Metrics Papers

Updated from current HFEPX corpus (Feb 27, 2026). 309 papers are grouped in this hub 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: Retrieval. 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 Feb 26, 2026.

Papers: 309 Last published: Feb 26, 2026 Global RSS Tag RSS
Cs.LGAutomatic Metrics

Research Narrative

Grounded narrative Model: deterministic-grounded Source: persisted

Updated from current HFEPX corpus (Feb 27, 2026). This page tracks 309 papers for CS.LG + Automatic Metrics Papers. Dominant protocol signals include automatic metrics, simulation environments, human evaluation, with frequent benchmark focus on Retrieval, MATH and metric focus on accuracy, cost. Use the grounded sections below to prioritize reproducible protocol choices, benchmark-matched comparisons, and judge-vs-human evaluation checks.

Why This Matters For Eval Research

Protocol Takeaways

Benchmark Interpretation

  • Retrieval appears in 4.2% of hub papers (13/309); use this cohort for benchmark-matched comparisons.
  • MATH appears in 2.9% of hub papers (9/309); use this cohort for benchmark-matched comparisons.

Metric Interpretation

  • accuracy is reported in 26.5% of hub papers (82/309); compare with a secondary metric before ranking methods.
  • cost is reported in 9.1% of hub papers (28/309); compare with a secondary metric before ranking methods.

Researcher Checklist

  • Close gap on Papers with explicit human feedback. Coverage is a replication risk (12% vs 45% target).
  • Close gap on Papers reporting quality controls. Coverage is a replication risk (4.9% vs 30% target).
  • Close gap on Papers naming benchmarks/datasets. Coverage is a replication risk (19.4% vs 35% target).
  • Maintain strength on Papers naming evaluation metrics. Coverage is strong (48.9% vs 35% target).
  • Close gap on Papers with known rater population. Coverage is a replication risk (10% vs 35% target).
  • Close gap on Papers with known annotation unit. Coverage is a replication risk (10.7% vs 35% target).

Papers with explicit human feedback

Coverage is a replication risk (12% vs 45% target).

Papers reporting quality controls

Coverage is a replication risk (4.9% vs 30% target).

Papers naming benchmarks/datasets

Coverage is a replication risk (19.4% vs 35% target).

Papers naming evaluation metrics

Coverage is strong (48.9% vs 35% target).

Papers with known rater population

Coverage is a replication risk (10% vs 35% target).

Papers with known annotation unit

Coverage is a replication risk (10.7% vs 35% target).

Suggested Reading Order

  1. 1. InnerQ: Hardware-aware Tuning-free Quantization of KV Cache for Large Language Models

    High citation traction makes this a useful baseline for method and protocol context.

  2. 2. Fine-Tuning Without Forgetting In-Context Learning: A Theoretical Analysis of Linear Attention Models

    High citation traction makes this a useful baseline for method and protocol context.

  3. 3. Modality Collapse as Mismatched Decoding: Information-Theoretic Limits of Multimodal LLMs

    High citation traction makes this a useful baseline for method and protocol context.

  4. 4. Assessing Deanonymization Risks with Stylometry-Assisted LLM Agent

    Adds automatic metrics for broader coverage within this hub.

  5. 5. MoDora: Tree-Based Semi-Structured Document Analysis System

    Adds automatic metrics for broader coverage within this hub.

  6. 6. NoRA: Breaking the Linear Ceiling of Low-Rank Adaptation via Manifold Expansion

    Adds automatic metrics for broader coverage within this hub.

  7. 7. OmniGAIA: Towards Native Omni-Modal AI Agents

    Adds automatic metrics for broader coverage within this hub.

Known Limitations

  • Only 4.9% of papers report quality controls; prioritize calibration/adjudication evidence.
  • Rater population is under-specified (10% coverage).
  • Narrative synthesis is grounded in metadata and abstracts only; full-paper implementation details are not parsed.

Research Utility Links

human_eval vs llm_as_judge

both=0, left_only=2, right_only=1

0 papers use both Human Eval and Llm As Judge.

human_eval vs automatic_metrics

both=2, left_only=0, right_only=307

2 papers use both Human Eval and Automatic Metrics.

llm_as_judge vs automatic_metrics

both=1, left_only=0, right_only=308

1 papers use both Llm As Judge and Automatic Metrics.

Benchmark Brief

Retrieval

Coverage: 13 papers (4.2%)

13 papers (4.2%) mention Retrieval.

Examples: MoDora: Tree-Based Semi-Structured Document Analysis System , Vectorizing the Trie: Efficient Constrained Decoding for LLM-based Generative Retrieval on Accelerators , Search-P1: Path-Centric Reward Shaping for Stable and Efficient Agentic RAG Training

Benchmark Brief

MATH

Coverage: 9 papers (2.9%)

9 papers (2.9%) mention MATH.

Examples: Overconfident Errors Need Stronger Correction: Asymmetric Confidence Penalties for Reinforcement Learning , Gradient Regularization Prevents Reward Hacking in Reinforcement Learning from Human Feedback and Verifiable Rewards , From Growing to Looping: A Unified View of Iterative Computation in LLMs

Benchmark Brief

DROP

Coverage: 8 papers (2.6%)

8 papers (2.6%) mention DROP.

Examples: Mitigating Structural Noise in Low-Resource S2TT: An Optimized Cascaded Nepali-English Pipeline with Punctuation Restoration , Subgroups of $U(d)$ Induce Natural RNN and Transformer Architectures , TFL: Targeted Bit-Flip Attack on Large Language Model

Metric Brief

accuracy

Coverage: 82 papers (26.5%)

82 papers (26.5%) mention accuracy.

Examples: InnerQ: Hardware-aware Tuning-free Quantization of KV Cache for Large Language Models , Assessing Deanonymization Risks with Stylometry-Assisted LLM Agent , MoDora: Tree-Based Semi-Structured Document Analysis System

Metric Brief

cost

Coverage: 28 papers (9.1%)

28 papers (9.1%) mention cost.

Examples: How Do Latent Reasoning Methods Perform Under Weak and Strong Supervision? , SWE-Protégé: Learning to Selectively Collaborate With an Expert Unlocks Small Language Models as Software Engineering Agents , JSAM: Privacy Straggler-Resilient Joint Client Selection and Incentive Mechanism Design in Differentially Private Federated Learning

Metric Brief

latency

Coverage: 15 papers (4.9%)

15 papers (4.9%) mention latency.

Examples: InnerQ: Hardware-aware Tuning-free Quantization of KV Cache for Large Language Models , Vectorizing the Trie: Efficient Constrained Decoding for LLM-based Generative Retrieval on Accelerators , SWE-Protégé: Learning to Selectively Collaborate With an Expert Unlocks Small Language Models as Software Engineering Agents

Most Cited In This Hub

Fast path to methods with the strongest citation traction in this scope.

Papers: InnerQ: Hardware-aware Tuning-free Quantization of KV Cache for Large Language Models , Fine-Tuning Without Forgetting In-Context Learning: A Theoretical Analysis of Linear Attention Models , Modality Collapse as Mismatched Decoding: Information-Theoretic Limits of Multimodal LLMs

Most Recent

Fast path to latest protocol changes and newly published evaluation setups.

Papers: InnerQ: Hardware-aware Tuning-free Quantization of KV Cache for Large Language Models , Fine-Tuning Without Forgetting In-Context Learning: A Theoretical Analysis of Linear Attention Models , Modality Collapse as Mismatched Decoding: Information-Theoretic Limits of Multimodal LLMs

Best Protocol Detail

Papers with explicit rater/unit metadata and quality-control signals for reproducibility.

Papers: An artificial intelligence framework for end-to-end rare disease phenotyping from clinical notes using large language models

Top Papers

Related Hubs