Skip to content
← Back to explorer

HFEPX Hub

CS.LG + Math Papers

Updated from current HFEPX corpus (Feb 27, 2026). 42 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: MATH. 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: 42 Last published: Feb 26, 2026 Global RSS Tag RSS
Cs.LGMath

Research Narrative

Grounded narrative Model: deterministic-grounded Source: persisted

Updated from current HFEPX corpus (Feb 27, 2026). This page tracks 42 papers for CS.LG + Math Papers. Dominant protocol signals include automatic metrics, simulation environments, with frequent benchmark focus on MATH, GSM8K and metric focus on accuracy, latency. 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

  • MATH appears in 21.4% of hub papers (9/42); use this cohort for benchmark-matched comparisons.
  • GSM8K appears in 14.3% of hub papers (6/42); use this cohort for benchmark-matched comparisons.

Metric Interpretation

  • accuracy is reported in 45.2% of hub papers (19/42); compare with a secondary metric before ranking methods.
  • latency is reported in 11.9% of hub papers (5/42); compare with a secondary metric before ranking methods.

Researcher Checklist

  • Close gap on Papers with explicit human feedback. Coverage is a replication risk (7.1% vs 45% target).
  • Close gap on Papers reporting quality controls. Coverage is a replication risk (11.9% vs 30% target).
  • Maintain strength on Papers naming benchmarks/datasets. Coverage is strong (47.6% vs 35% target).
  • Maintain strength on Papers naming evaluation metrics. Coverage is strong (64.3% vs 35% target).
  • Close gap on Papers with known rater population. Coverage is a replication risk (7.1% vs 35% target).
  • Close gap on Papers with known annotation unit. Coverage is a replication risk (9.5% vs 35% target).

Papers with explicit human feedback

Coverage is a replication risk (7.1% vs 45% target).

Papers reporting quality controls

Coverage is a replication risk (11.9% vs 30% target).

Papers naming benchmarks/datasets

Coverage is strong (47.6% vs 35% target).

Papers naming evaluation metrics

Coverage is strong (64.3% vs 35% target).

Papers with known rater population

Coverage is a replication risk (7.1% vs 35% target).

Papers with known annotation unit

Coverage is a replication risk (9.5% vs 35% target).

Suggested Reading Order

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

    Start here for detailed protocol reporting, including rater and quality-control evidence.

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

    Start here for detailed protocol reporting, including rater and quality-control evidence.

  3. 3. Duel-Evolve: Reward-Free Test-Time Scaling via LLM Self-Preferences

    Start here for detailed protocol reporting, including rater and quality-control evidence.

  4. 4. Overconfident Errors Need Stronger Correction: Asymmetric Confidence Penalties for Reinforcement Learning

    Adds automatic metrics for broader coverage within this hub.

  5. 5. Black-Box Reliability Certification for AI Agents via Self-Consistency Sampling and Conformal Calibration

    Adds automatic metrics for broader coverage within this hub.

  6. 6. Equitable Evaluation via Elicitation

    Adds automatic metrics for broader coverage within this hub.

  7. 7. Aletheia tackles FirstProof autonomously

    Adds automatic metrics for broader coverage within this hub.

  8. 8. Why Pass@k Optimization Can Degrade Pass@1: Prompt Interference in LLM Post-training

    Adds automatic metrics for broader coverage within this hub.

Known Limitations

  • Only 11.9% of papers report quality controls; prioritize calibration/adjudication evidence.
  • Rater population is under-specified (7.1% coverage).
  • Narrative synthesis is grounded in metadata and abstracts only; full-paper implementation details are not parsed.

Research Utility Links

automatic_metrics vs simulation_env

both=0, left_only=39, right_only=3

0 papers use both Automatic Metrics and Simulation Env.

Benchmark Brief

MATH

Coverage: 9 papers (21.4%)

9 papers (21.4%) 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

GSM8K

Coverage: 6 papers (14.3%)

6 papers (14.3%) mention GSM8K.

Examples: InnerQ: Hardware-aware Tuning-free Quantization of KV Cache for Large Language Models , Black-Box Reliability Certification for AI Agents via Self-Consistency Sampling and Conformal Calibration , Pyramid MoA: A Probabilistic Framework for Cost-Optimized Anytime Inference

Benchmark Brief

ARC

Coverage: 3 papers (7.1%)

3 papers (7.1%) mention ARC.

Examples: Recursive Concept Evolution for Compositional Reasoning in Large Language Models , Weight space Detection of Backdoors in LoRA Adapters , Improving Denoising Diffusion Models via Simultaneous Estimation of Image and Noise

Metric Brief

accuracy

Coverage: 19 papers (45.2%)

19 papers (45.2%) mention accuracy.

Examples: InnerQ: Hardware-aware Tuning-free Quantization of KV Cache for Large Language Models , Duel-Evolve: Reward-Free Test-Time Scaling via LLM Self-Preferences , Overconfident Errors Need Stronger Correction: Asymmetric Confidence Penalties for Reinforcement Learning

Metric Brief

latency

Coverage: 5 papers (11.9%)

5 papers (11.9%) mention latency.

Examples: InnerQ: Hardware-aware Tuning-free Quantization of KV Cache for Large Language Models , Why Pass@k Optimization Can Degrade Pass@1: Prompt Interference in LLM Post-training , Pyramid MoA: A Probabilistic Framework for Cost-Optimized Anytime Inference

Metric Brief

calibration

Coverage: 3 papers (7.1%)

3 papers (7.1%) mention calibration.

Examples: Pyramid MoA: A Probabilistic Framework for Cost-Optimized Anytime Inference , Classification errors distort findings in automated speech processing: examples and solutions from child-development research , Humanity's Last Exam

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 , NoRA: Breaking the Linear Ceiling of Low-Rank Adaptation via Manifold Expansion , Duel-Evolve: Reward-Free Test-Time Scaling via LLM Self-Preferences

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 , NoRA: Breaking the Linear Ceiling of Low-Rank Adaptation via Manifold Expansion , Duel-Evolve: Reward-Free Test-Time Scaling via LLM Self-Preferences

Best Protocol Detail

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

Papers: InnerQ: Hardware-aware Tuning-free Quantization of KV Cache for Large Language Models , NoRA: Breaking the Linear Ceiling of Low-Rank Adaptation via Manifold Expansion , Duel-Evolve: Reward-Free Test-Time Scaling via LLM Self-Preferences

Top Papers

Related Hubs