Skip to content
OpenTrain AIFor AI Companies

HFEPX · Eval paper review

HARPO: Hallucination-Aware Reinforcement Learning for Faithful and Creative Language Generation

Tiezheng Yu, Yuxin Jiang, Jinpeng Li, Shuning Sun +3 more

Published

Oct 2, 2026

Citations

0

Trust level

Low

Usefulness score

5/100 (Low)

Extraction confidence

45% (Low)

Derived from extracted protocol signals and abstract evidence.

Rater population

Not reported

Signals refreshed

Oct 2, 2026

Open arXiv Find implementation Shortlist (0) RSS feed

Tags

Automatic Metrics

Should you rely on this paper?

This paper is adjacent to HFEPX scope and is best used for background context, not as a primary protocol reference.

Use this as background context only. Do not make protocol decisions from this page alone.

Best use

Background context only

Use if you need

A benchmark-and-metrics comparison anchor.

What to verify

Validate the evaluation procedure and quality controls in the full paper before operational use.

Main weakness

This paper looks adjacent to evaluation work, but not like a strong protocol reference.

Human feedback signal
Not explicit
Not explicit in abstract metadata
Evaluation signal
Detected
Eval setup described
Usefulness for eval research
5/100
Adjacent candidate

Treat as adjacent context, not a core eval-method reference.

If you are doing eval pipeline work, start here

Human Eval Papers LLM-as-Judge Papers Pairwise Preference Papers Tool-Use Evaluation Search

Abstract

Large Language Models (LLMs) are prone to generating hallucinated content, which compromises their reliability in knowledge-intensive tasks. To address this challenge without sacrificing creativity, we propose HARPO, a reinforcement learning framework designed to jointly optimize faithfulness and creativity. HARPO incorporates a Hallucination-Aware Generative Reward Model (HA-GRM), trained via verifiable feedback, to assess both faithfulness and writing quality. A Selective Activation Mechanism (SAM) activates writing rewards only for outputs judged hallucination-free by HA-GRM, while a data curriculum progressively shifts training from creative writing to hallucination-centric tasks. On RAGTruth, our Qwen3-4B-based HA-GRM achieves a response-level F1 score of 78.08%, compared with 66.37% for the supervised fine-tuning baseline. Experiments on Qwen2.5 and Qwen3 models from 1.7B to 8B parameters show improvements in both faithful generation and writing quality. On Qwen3-4B, HARPO reduces the HA-GRM-judged hallucination rate on MultiHopRAG from 3.29% to 1.02%, while increasing the Arena-Hard-v2.0 creative-writing score from 16.95% to 27.54%.

What we could verify

These are the protocol signals we could actually recover from the available paper metadata. Use them to decide whether this paper is worth deeper reading.

Human Feedback Types

missing

None explicit

No explicit feedback protocol extracted.

"Large Language Models (LLMs) are prone to generating hallucinated content, which compromises their reliability in knowledge-intensive tasks."

Evaluation Modes

partial

Automatic Metrics

Includes extracted eval setup.

"Large Language Models (LLMs) are prone to generating hallucinated content, which compromises their reliability in knowledge-intensive tasks."

Quality Controls

missing

Not reported

No explicit QC controls found.

"Large Language Models (LLMs) are prone to generating hallucinated content, which compromises their reliability in knowledge-intensive tasks."

Benchmarks / Datasets

partial

LMSYS Chatbot Arena, Arena Hard, RAGTruth

Useful for quick benchmark comparison.

"On RAGTruth, our Qwen3-4B-based HA-GRM achieves a response-level F1 score of 78.08%, compared with 66.37% for the supervised fine-tuning baseline."

Reported Metrics

partial

F1, Hallucination rate, Faithfulness

Useful for evaluation criteria comparison.

"To address this challenge without sacrificing creativity, we propose HARPO, a reinforcement learning framework designed to jointly optimize faithfulness and creativity."

Benchmarks and datasets

LMSYS Chatbot ArenaArena-HardRAGTruth

Reported metrics

f1hallucination ratefaithfulness
Human feedback details
Uses human feedback
No
Feedback types
None
Rater population
Not reported
Unit of annotation
Scalar (inferred)
Expertise required
General
Evaluation details
Evaluation modes
Automatic Metrics
Agentic eval
None
Quality controls
Not reported
Evidence quality
Low
Use this page as
Background context only

Research brief

Metadata summary

Large Language Models (LLMs) are prone to generating hallucinated content, which compromises their reliability in knowledge-intensive tasks.

Based on abstract + metadata only. Check the source paper before making high-confidence protocol decisions.

Key takeaways

  • Large Language Models (LLMs) are prone to generating hallucinated content, which compromises their reliability in knowledge-intensive tasks.
  • To address this challenge without sacrificing creativity, we propose HARPO, a reinforcement learning framework designed to jointly optimize faithfulness and creativity.
  • HARPO incorporates a Hallucination-Aware Generative Reward Model (HA-GRM), trained via verifiable feedback, to assess both faithfulness and writing quality.

Researcher actions

  • Compare this paper against nearby papers in the same arXiv category before using it for protocol decisions.
  • Validate inferred eval signals (Automatic metrics) against the full paper.
  • Use related-paper links to find stronger protocol-specific references.

Caveats

  • Generated from abstract + metadata only; no PDF parsing.
  • Signals below are heuristic and may miss details reported outside the abstract.

Recommended queries

F1

Contribution summary

  • To address this challenge without sacrificing creativity, we propose HARPO, a reinforcement learning framework designed to jointly optimize faithfulness and creativity.
  • A Selective Activation Mechanism (SAM) activates writing rewards only for outputs judged hallucination-free by HA-GRM, while a data curriculum progressively shifts training from creative writing to hallucination-centric tasks.
  • On Qwen3-4B, HARPO reduces the HA-GRM-judged hallucination rate on MultiHopRAG from 3.29% to 1.02%, while increasing the Arena-Hard-v2.0 creative-writing score from 16.95% to 27.54%.

Why it matters for eval

  • A Selective Activation Mechanism (SAM) activates writing rewards only for outputs judged hallucination-free by HA-GRM, while a data curriculum progressively shifts training from creative writing to hallucination-centric tasks.
  • On Qwen3-4B, HARPO reduces the HA-GRM-judged hallucination rate on MultiHopRAG from 3.29% to 1.02%, while increasing the Arena-Hard-v2.0 creative-writing score from 16.95% to 27.54%.

Researcher checklist

  • Human feedback protocol is explicit

    No explicit human feedback protocol detected.

  • Evaluation mode is explicit

    Detected: Automatic Metrics

  • Quality control reporting appears

    No calibration/adjudication/IAA control explicitly detected.

  • Benchmark or dataset anchors are present

    Detected: LMSYS Chatbot Arena, Arena-Hard, RAGTruth

  • Metric reporting is present

    Detected: f1, hallucination rate, faithfulness