Role-playing with large language models is fundamentally a *session-level* task, requiring agents to sustain character identity and interaction quality across extended multi-turn conversations. Yet existing evaluation and optimization methods remain largely turn-level, failing to capture long-horizon quality. We propose **DynSess**, a unified session-level framework for role-playing agents. **DynSess-Eval** scores complete dialogue sessions via rubrics targeting long-horizon behaviors. Leveraging its session-level rewards, we construct high-quality training trajectories through multi-turn lookahead search and train **DynSess-Character** with two complementary variants: DSPO (off-policy) and GSRPO (on-policy). Experiments show that DynSess-Eval aligns with human judgments substantially better than prior evaluators, and blind human evaluation further shows that DynSess-Character matches the strongest character model despite using substantially fewer parameters, while maintaining strong role consistency and interactive ability. The dataset and code are publicly available in https://github.com/Collab-Gen/DynSess.

While role-playing agents excel in short-term interactions, long-term conversations overwhelm context windows, motivating external memory frameworks. Current systems typically rely on persona-agnostic summarization, which records facts without persona-specific interpretation, yielding generic responses that compromise persona fidelity. To bridge this gap, we introduce RoleMemo, a dataset featuring four reasoning tasks where the factual fragments must be interpreted through the persona to reach the correct answer. Evaluation on RoleMemo exposes critical limitations of persona-agnostic frameworks. We thus propose DualMem, which decouples memory into two streams: factual cognition and persona-conditioned insight. Trained through Supervised Fine-Tuning (SFT) and Reinforcement Learning (RL), our framework with a 4B-parameter model outperforms zero-shot persona-agnostic frameworks powered by DeepSeek-V3.2 for sustained persona fidelity. Our resources are available at https://github.com/Collab-Gen/RoleMemo.

Complex decision-making often requires agents to progress through intermediate tasks rather than solve the final target directly. Existing LLM self-refinement methods typically iterate on a fixed target, while curriculum-learning methods often rely on hand-designed schedules, training-time optimization, or domain-specific difficulty metrics. To smooth the learning curve through adaptive curriculum design, we introduce EvoCurr, a general inference-time framework that co-evolves curricula and executable policies. A Designer proposes verifiable intermediate tasks from the latest accepted progress and recent failures, while a Solver generates or trains policies for these tasks. A task-policy pair is accepted only when it passes hard feasibility checks and reaches a specified performance threshold; meanwhile, an accepted-floor constraint preserves the latest verified checkpoint and prevents failed harder attempts from overwriting mastered behavior. We evaluate EvoCurr across code-as-policy and closed-loop MARL settings, using StarCraft II super-late-game micromanagement as our hardest stress test. EvoCurr demonstrates effective micromanagement across three LLM backbones without LLM parameter training. Improvements in additional experiments on Flatland, POGEMA, Overcooked, and LiveCodeBench Hard demonstrate the generality of EvoCurr.

With the rapid development of large language models (LLMs), role-playing dialogue systems have become increasingly popular for immersive and personalized human–AI interaction. However, in real-world role-playing scenarios, users often provide lazy queries—brief and ambiguous responses such as "oh," "hmm," or "?"—which lack explicit intent and emotional cues, making it difficult for role-playing LLMs to generate engaging replies. Existing query rewriting methods mainly target functional tasks such as information retrieval and are not tailored to role-playing dialogue, where rewrite quality depends on role-playing-specific factors such as character consistency and personalized user preferences. To address this, we first propose a listwise Generative Reward Model (GRM) trained from real user preference data collected through online regeneration behavior, which provides fine-grained process and final preference signals for training a Lazy Query Rewriter. Using rejection fine-tuning and reinforcement learning, the LQ-Rewriter learns to transform underspecified user inputs into richer expressions while preserving user intent and dialogue context. Experiments on real-world dialogue logs from a mainstream Chinese AI-powered romantic role-playing platform show that our GRM substantially outperforms general-purpose LLM judges. Moreover, our LQ-Rewriter achieves the highest GRM-based and human preference win rates among all rewriting baselines, and brings tangible improvements in online deployment.

Personality is an important concept in psychology that reflects individual differences in thinking and behavior, and has significant applications across various fields. Most existing personality analysis methods address this issue at the bag level, treating the entire corpus gathered from one individual as a single unit for classification. However, this paradigm presents several challenges. From the data perspective, collecting a large corpus for each individual and performing comprehensive annotations pose significant difficulties in both data collection and labeling. On the application side, concentrating on classifying the entire corpus limits its applicability in more common single-instance scenarios. To address these issues, we propose a new task paradigm in text-based personality representation learning. Specifically, we construct a triplet personality trend comparison dataset to learn single-sentence personality embeddings with desirable metric properties. This approach removes the traditional constraints on data sources, facilitating dataset expansion, and can leverage the transfer capabilities of embeddings to easily adapt to various downstream tasks. Our experiments show that the learned embeddings significantly boost performance by a relative 10% across various applications, including personality detection, personality retrieval, and emotion translation prediction. The code and dataset are available at https://github.com/zjutangk/PTCD.

Cognitive Restructuring (CR) uses multi-turn dialogue to identify and restructure one’s negative thoughts, arising from mental health issues, into more helpful and positive ones. Clinician shortage and stigma urge the development of human-LLM interactive psychotherapy for CR. Yet, effectively implementing CR is hindered by entrenched cognitive distortions, emotional resistance, and individual differences, which existing works have not overcome. To bridge this gap, we propose CRDial, a novel framework that structures CR as theory-grounded multi-stage multi-turn dialogue, integrating multi-aspect supportive strategies for emotional management and a multi-channel loop mechanism to account for diverse individual distortions. With CRDial, we distill Crisp, a large-scale and high-quality bilingual dialogue dataset, from LLM. We then train Crispers, Crisp-based conversational LLMs for CR, at 7B and 14B scales. Extensive human studies show the superiority of Crispers in pointwise, pairwise, and intervention evaluations.

The use of modern Large Language Models (LLMs) as chatbots still has some problems such as hallucinations and lack of empathy. Identifying these issues can help improve chatbot performance. The community has been continually iterating on reference-free dialogue evaluation methods based on large language models (LLMs) that can be readily applied. However, many of these LLM-based metrics require selecting specific datasets and developing specialized training tasks for different evaluation dimensions (e.g., coherence, informative). The developing step can be time-consuming and may need to be repeated for new evaluation dimensions. To enable efficient and flexible adaptation to diverse needs of dialogue evaluation, we propose a dimension-agnostic scoring method that leverages the in-context learning (ICL) capability of LLMs to learn from human scoring to the fullest extent. Our method has three key features. To begin with, rather than manual prompt crafting, we propose automatically generating prompts, allowing the LLM to observe human labels and summarize the most suitable prompt. Additionally, since the LLM has a token limit and ICL is sensitive to demonstration variations, we train a selector to finely customize demonstrations and prompts for each dialogue input. Finally, during inference, we propose to request the LLM multiple times with a subgraph of demonstrations and prompts that are diverse and suitable to maximize ICL from various human scoring. We validate the efficacy of our method on five datasets, even with a small amount of annotated data, our method outperforms all strong baselines. Code is available at https://github.com/iamlxb3/EMNLP2023-ADOROR.