RoleJudge is a multidimensional evaluation framework for speech-character alignment in audio LLMs, backed by the RoleChat dataset and multi-stage RL training with standard alignment to reduce reward issues.
ChatHaruhi: Reviving Anime Character in Reality via Large Language Model
12 Pith papers cite this work, alongside 11 external citations. Polarity classification is still indexing.
12
Pith papers citing it
11
external citations · Pith
abstract
Role-playing chatbots built on large language models have drawn interest, but better techniques are needed to enable mimicking specific fictional characters. We propose an algorithm that controls language models via an improved prompt and memories of the character extracted from scripts. We construct ChatHaruhi, a dataset covering 32 Chinese / English TV / anime characters with over 54k simulated dialogues. Both automatic and human evaluations show our approach improves role-playing ability over baselines. Code and data are available at https://github.com/LC1332/Chat-Haruhi-Suzumiya .
citation-role summary
background 2
citation-polarity summary
roles
background 2polarities
background 2representative citing papers
HER trains LLMs on reverse-engineered reasoning data and human preference rewards to improve cognitive persona simulation, reporting 30-point gains on CoSER and 15% on Minimax benchmarks over Qwen3-32B.
AudioRole provides 1M+ character-grounded audio-text dialogues from TV series plus ARP-Eval to train and measure audio role-playing models, with ARP-Model showing 0.31 acoustic and 0.36 content personalization scores.
A CPM-grounded multi-agent system extracts dialogue triggers, appraises them on relevance/implication/coping/norms, and updates a persona’s latent multi-emotion state more coherently than standard prompting baselines.
Psy-CoT decomposes reasoning into Interaction Perception, Psychological Empathy, and Logical Construction while RAPO asymmetrically weights role-specific tokens during policy optimization, outperforming prior CoT and GRPO baselines on role-playing benchmarks.
New benchmark RoleCDE reveals LLMs exhibit role value decoupling under conflicts and demonstrates mitigation via targeted fine-tuning.
DynSess supplies session-level rubrics for dialogue evaluation and uses the resulting rewards to train lighter role-playing agents via multi-turn lookahead search and DSPO/GSRPO optimization that match stronger baselines on human judgments.
PersonaArena is a dynamic simulation framework that constructs persona banks from social data and uses multi-agent debating judges to evaluate and enhance persona-level role-playing in LLMs.
ToxPrune prunes toxic subwords from BPE tokenizers in LLMs to mitigate toxic dialogue responses and improve diversity on both toxic and non-toxic models.
A pipeline using causal language modeling and synthetic instruction-tuning personalizes LLMs to replicate individual Hebrew judges' reasoning, outperforming baselines on similarity metrics with outputs indistinguishable from human judges.
A survey proposing a three-level capability taxonomy (L1 Predictor, L2 Simulator, L3 Evolver) for world models across physical, digital, social, and scientific domains.
A systematic review of memory designs, evaluation methods, applications, limitations, and future directions for LLM-based agents.
citing papers explorer
-
Character Beyond Speech: Leveraging Role-Playing Evaluation in Audio Large Language Models via Reinforcement Learning
RoleJudge is a multidimensional evaluation framework for speech-character alignment in audio LLMs, backed by the RoleChat dataset and multi-stage RL training with standard alignment to reduce reward issues.
-
HER: Human-like Reasoning and Reinforcement Learning for LLM Role-playing
HER trains LLMs on reverse-engineered reasoning data and human preference rewards to improve cognitive persona simulation, reporting 30-point gains on CoSER and 15% on Minimax benchmarks over Qwen3-32B.
-
AudioRole: An Audio Dataset for Character Role-Playing in Large Language Models
AudioRole provides 1M+ character-grounded audio-text dialogues from TV series plus ARP-Eval to train and measure audio role-playing models, with ARP-Model showing 0.31 acoustic and 0.36 content personalization scores.
-
From Triggers to Emotions: A CPM-Grounded Appraisal Multi-Agent for Dynamic Emotional Evolution in Persona-Based Dialogue
A CPM-grounded multi-agent system extracts dialogue triggers, appraises them on relevance/implication/coping/norms, and updates a persona’s latent multi-emotion state more coherently than standard prompting baselines.
-
Improving General Role-Playing Agents via Psychology-Grounded Reasoning and Role-Aware Policy Optimization
Psy-CoT decomposes reasoning into Interaction Perception, Psychological Empathy, and Logical Construction while RAPO asymmetrically weights role-specific tokens during policy optimization, outperforming prior CoT and GRPO baselines on role-playing benchmarks.
-
RoleCDE:Benchmarking and Mitigating Role-Alignment Trade-offs in Role-Playing Agents
New benchmark RoleCDE reveals LLMs exhibit role value decoupling under conflicts and demonstrates mitigation via targeted fine-tuning.
-
DynSess: Dynamic Session-Level Evaluation and Optimization Framework for Role-Playing Agents
DynSess supplies session-level rubrics for dialogue evaluation and uses the resulting rewards to train lighter role-playing agents via multi-turn lookahead search and DSPO/GSRPO optimization that match stronger baselines on human judgments.
-
PersonaArena: Dynamic Simulation for Evaluating and Enhancing Persona-Level Role-Playing in Large Language Models
PersonaArena is a dynamic simulation framework that constructs persona banks from social data and uses multi-agent debating judges to evaluate and enhance persona-level role-playing in LLMs.
-
Toxic Subword Pruning for Dialogue Response Generation on Large Language Models
ToxPrune prunes toxic subwords from BPE tokenizers in LLMs to mitigate toxic dialogue responses and improve diversity on both toxic and non-toxic models.
-
JudgeMeNot: Personalizing Large Language Models to Emulate Judicial Reasoning in Hebrew
A pipeline using causal language modeling and synthetic instruction-tuning personalizes LLMs to replicate individual Hebrew judges' reasoning, outperforming baselines on similarity metrics with outputs indistinguishable from human judges.
-
Agentic World Modeling: Foundations, Capabilities, Laws, and Beyond
A survey proposing a three-level capability taxonomy (L1 Predictor, L2 Simulator, L3 Evolver) for world models across physical, digital, social, and scientific domains.
-
A Survey on the Memory Mechanism of Large Language Model based Agents
A systematic review of memory designs, evaluation methods, applications, limitations, and future directions for LLM-based agents.