Quantifying and Optimizing Global Faithfulness in Persona-driven Role-playing

• URL: http://arxiv.org/abs/2405.07726v2
• Date: Wed, 16 Oct 2024 02:54:48 GMT
• Title: Quantifying and Optimizing Global Faithfulness in Persona-driven Role-playing
• Authors: Letian Peng, Jingbo Shang,
• Abstract summary: Persona-driven role-playing (PRP) aims to build AI characters that can respond to user queries by faithfully sticking with all persona statements. This paper presents a pioneering exploration to quantify PRP faithfulness as a fine-grained and explainable criterion, which also serves as a reliable reference for optimization.
• Score: 37.92922713921964
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Persona-driven role-playing (PRP) aims to build AI characters that can respond to user queries by faithfully sticking with all persona statements. Unfortunately, existing faithfulness criteria for PRP are limited to coarse-grained LLM-based scoring without a clear definition or formulation. This paper presents a pioneering exploration to quantify PRP faithfulness as a fine-grained and explainable criterion, which also serves as a reliable reference for optimization. Our criterion first discriminates persona statements into active and passive constraints by identifying the query-statement relevance. Then, we incorporate all constraints following the principle that the AI character's response should be (a) entailed by active (relevant) constraints and (b) not contradicted by passive (irrelevant) constraints. We translate this principle mathematically into a novel Active-Passive-Constraint (APC) score, a constraint-wise sum of natural language inference (NLI) scores weighted by relevance scores. In practice, we build the APC scoring system by symbolically distilling small discriminators from GPT-4 for efficiency. We validate the quality of the APC score against human evaluation based on example personas with tens of statements, and the results show a high correlation. We further leverage it as a reward system in direct preference optimization (DPO) for better AI characters. Our experiments offer a fine-grained and explainable comparison between existing PRP techniques, revealing their advantages and limitations. We further find APC-based DPO to be one of the most competitive techniques for sticking with all constraints and can be well incorporated with other techniques. We then extend the scale of the experiments to real persons with hundreds of statements and reach a consistent conclusion.

Related papers

• Conformal Prediction Beyond the Seen: A Missing Mass Perspective for Uncertainty Quantification in Generative Models [20.810300785340072]
  Conformal Prediction with Query Oracle (CPQ) is a framework characterizing the optimal interplay between these objectives.<n>Our algorithm is built on two core principles: one governs the optimal query policy, and the other defines the optimal mapping from queried samples to prediction sets.
  arXiv  Detail & Related papers  (2025-06-05T18:26:14Z)
• RAG-Zeval: Towards Robust and Interpretable Evaluation on RAG Responses through End-to-End Rule-Guided Reasoning [64.46921169261852]
  RAG-Zeval is a novel end-to-end framework that formulates faithfulness and correctness evaluation as a rule-guided reasoning task.<n>Our approach trains evaluators with reinforcement learning, facilitating compact models to generate comprehensive and sound assessments.<n>Experiments demonstrate RAG-Zeval's superior performance, achieving the strongest correlation with human judgments.
  arXiv  Detail & Related papers  (2025-05-28T14:55:33Z)
• Charting the Parrot's Song: A Maximum Mean Discrepancy Approach to Measuring AI Novelty, Originality, and Distinctiveness [0.2209921757303168]
  This paper introduces a robust, quantitative methodology to measure distributional differences between generative processes. By comparing entire output distributions rather than conducting pairwise similarity checks, our approach directly contrasts creative processes. This research provides courts and policymakers with a computationally efficient, legally relevant tool to quantify AI novelty.
  arXiv  Detail & Related papers  (2025-04-11T11:15:26Z)
• Pareto Set Identification With Posterior Sampling [14.121842087273167]
  This paper studies PSI in the transductive linear setting with potentially correlated objectives. We propose the PSIPS algorithm that deals simultaneously with structure and correlation without paying the computational cost of existing oracle-based algorithms.
  arXiv  Detail & Related papers  (2024-11-07T18:15:38Z)
• Abstract Reward Processes: Leveraging State Abstraction for Consistent Off-Policy Evaluation [20.663398371026194]
  We introduce STAR, a framework for off-policy evaluation that encompasses a broad range of estimators. We empirically demonstrate that estimators within STAR outperform existing methods. The best STAR estimator outperforms baselines in all twelve cases studied.
  arXiv  Detail & Related papers  (2024-10-03T03:19:43Z)
• $i$REPO: $i$mplicit Reward Pairwise Difference based Empirical Preference Optimization [12.266207199002604]
  Large Language Models (LLM) can sometimes produce outputs that deviate from human expectations. We propose a novel framework named $i$REPO, which utilizes implicit Reward pairwise difference regression for Empirical Preference Optimization. We show that $i$REPO effectively achieves self-alignment using soft-label, self-generated responses and the logit of empirical AI annotators.
  arXiv  Detail & Related papers  (2024-05-24T05:42:11Z)
• Query Performance Prediction using Relevance Judgments Generated by Large Language Models [53.97064615557883]
  We propose a QPP framework using automatically generated relevance judgments (QPP-GenRE) QPP-GenRE decomposes QPP into independent subtasks of predicting relevance of each item in a ranked list to a given query. This allows us to predict any IR evaluation measure using the generated relevance judgments as pseudo-labels.
  arXiv  Detail & Related papers  (2024-04-01T09:33:05Z)
• Probabilistic Offline Policy Ranking with Approximate Bayesian Computation [4.919605764492689]
  It is essential to compare and rank candidate policies offline before real-world deployment for safety and reliability. We present Probabilistic Offline Policy Ranking (POPR), a framework to address OPR problems. POPR does not rely on value estimation, and the derived performance posterior can be used to distinguish candidates in worst, best, and average-cases.
  arXiv  Detail & Related papers  (2023-12-17T05:22:44Z)
• Pairwise Proximal Policy Optimization: Harnessing Relative Feedback for LLM Alignment [37.52249093928251]
  This paper proposes a new framework, reinforcement learning with relative feedback, and a novel trajectory-wise policy gradient algorithm. We show theoretically that P3O is invariant to equivalent rewards and avoids the complexity of PPO. Empirical evaluations demonstrate that P3O outperforms PPO in the KL-Reward trade-off and can align with human preferences as well as or better than prior methods.
  arXiv  Detail & Related papers  (2023-09-30T01:23:22Z)
• Secrets of RLHF in Large Language Models Part I: PPO [81.01936993929127]
  Large language models (LLMs) have formulated a blueprint for the advancement of artificial general intelligence. reinforcement learning with human feedback (RLHF) emerges as the pivotal technological paradigm underpinning this pursuit. In this report, we dissect the framework of RLHF, re-evaluate the inner workings of PPO, and explore how the parts comprising PPO algorithms impact policy agent training.
  arXiv  Detail & Related papers  (2023-07-11T01:55:24Z)
• Policy learning "without" overlap: Pessimism and generalized empirical Bernstein's inequality [94.89246810243053]
  This paper studies offline policy learning, which aims at utilizing observations collected a priori to learn an optimal individualized decision rule. Existing policy learning methods rely on a uniform overlap assumption, i.e., the propensities of exploring all actions for all individual characteristics must be lower bounded. We propose Pessimistic Policy Learning (PPL), a new algorithm that optimize lower confidence bounds (LCBs) instead of point estimates.
  arXiv  Detail & Related papers  (2022-12-19T22:43:08Z)
• B-Pref: Benchmarking Preference-Based Reinforcement Learning [84.41494283081326]
  We introduce B-Pref, a benchmark specially designed for preference-based RL. A key challenge with such a benchmark is providing the ability to evaluate candidate algorithms quickly. B-Pref alleviates this by simulating teachers with a wide array of irrationalities.
  arXiv  Detail & Related papers  (2021-11-04T17:32:06Z)
• An Information Bottleneck Approach for Controlling Conciseness in Rationale Extraction [84.49035467829819]
  We show that it is possible to better manage this trade-off by optimizing a bound on the Information Bottleneck (IB) objective. Our fully unsupervised approach jointly learns an explainer that predicts sparse binary masks over sentences, and an end-task predictor that considers only the extracted rationale.
  arXiv  Detail & Related papers  (2020-05-01T23:26:41Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.