Large Language Model Agents Are Not Always Faithful Self-Evolvers

Self-evolving large language model (LLM) agents continually improve by accumulating and reusing past experience, yet it remains unclear whether they faithfully rely on that experience to guide their behavior. We present the first systematic investigation of experience faithfulness, the causal dependence of an agent's decisions on the experience it is given, in self-evolving LLM agents. Using controlled causal interventions on both raw and condensed forms of experience, we comprehensively evaluate four representative frameworks across 10 LLM backbones and 9 environments. Our analysis uncovers a striking asymmetry: while agents consistently depend on raw experience, they often disregard or misinterpret condensed experience, even when it is the only experience provided. This gap persists across single- and multi-agent configurations and across backbone scales. We trace its underlying causes to three factors: the semantic limitations of condensed content, internal processing biases that suppress experience, and task regimes where pretrained priors already suffice. These findings challenge prevailing assumptions about self-evolving methods and underscore the need for more faithful and reliable approaches to experience integration.

TEA-Bench: A Systematic Benchmarking of Tool-enhanced Emotional Support Dialogue Agent

Emotional Support Conversation requires not only affective expression but also grounded instrumental support to provide trustworthy guidance. However, existing ESC systems and benchmarks largely focus on affective support in text-only settings, overlooking how external tools can enable factual grounding and reduce hallucination in multi-turn emotional support. We introduce TEA-Bench, the first interactive benchmark for evaluating tool-augmented agents in ESC, featuring realistic emotional scenarios, an MCP-style tool environment, and process-level metrics that jointly assess the quality and factual grounding of emotional support. Experiments on nine LLMs show that tool augmentation generally improves emotional support quality and reduces hallucination, but the gains are strongly capacity-dependent: stronger models use tools more selectively and effectively, while weaker models benefit only marginally. We further release TEA-Dialog, a dataset of tool-enhanced ESC dialogues, and find that supervised fine-tuning improves in-distribution support but generalizes poorly. Our results underscore the importance of tool use in building reliable emotional support agents.

When Personalization Legitimizes Risks: Uncovering Safety Vulnerabilities in Personalized Dialogue Agents

Long-term memory enables large language model (LLM) agents to support personalized and sustained interactions. However, most work on personalized agents prioritizes utility and user experience, treating memory as a neutral component and largely overlooking its safety implications. In this paper, we reveal intent legitimation, a previously underexplored safety failure in personalized agents, where benign personal memories bias intent inference and cause models to legitimize inherently harmful queries. To study this phenomenon, we introduce PS-Bench, a benchmark designed to identify and quantify intent legitimation in personalized interactions. Across multiple memory-augmented agent frameworks and base LLMs, personalization increases attack success rates by 15.8%-243.7% relative to stateless baselines. We further provide mechanistic evidence for intent legitimation from internal representations space, and propose a lightweight detection-reflection method that effectively reduces safety degradation. Overall, our work provides the first systematic exploration and evaluation of intent legitimation as a safety failure mode that naturally arises from benign, real-world personalization, highlighting the importance of assessing safety under long-term personal context. WARNING: This paper may contain harmful content.

OP-Bench: Benchmarking Over-Personalization for Memory-Augmented Personalized Conversational Agents

Memory-augmented conversational agents enable personalized interactions using long-term user memory and have gained substantial traction. However, existing benchmarks primarily focus on whether agents can recall and apply user information, while overlooking whether such personalization is used appropriately. In fact, agents may overuse personal information, producing responses that feel forced, intrusive, or socially inappropriate to users. We refer to this issue as \emph{over-personalization}. In this work, we formalize over-personalization into three types: Irrelevance, Repetition, and Sycophancy, and introduce \textbf{OP-Bench} a benchmark of 1,700 verified instances constructed from long-horizon dialogue histories. Using \textbf{OP-Bench}, we evaluate multiple large language models and memory-augmentation methods, and find that over-personalization is widespread when memory is introduced. Further analysis reveals that agents tend to retrieve and over-attend to user memories even when unnecessary. To address this issue, we propose \textbf{Self-ReCheck}, a lightweight, model-agnostic memory filtering mechanism that mitigates over-personalization while preserving personalization performance. Our work takes an initial step toward more controllable and appropriate personalization in memory-augmented dialogue systems.

STAR-S: Improving Safety Alignment through Self-Taught Reasoning on Safety Rules

Defending against jailbreak attacks is crucial for the safe deployment of Large Language Models (LLMs). Recent research has attempted to improve safety by training models to reason over safety rules before responding. However, a key issue lies in determining what form of safety reasoning effectively defends against jailbreak attacks, which is difficult to explicitly design or directly obtain. To address this, we propose \textbf{STAR-S} (\textbf{S}elf-\textbf{TA}ught \textbf{R}easoning based on \textbf{S}afety rules), a framework that integrates the learning of safety rule reasoning into a self-taught loop. The core of STAR-S involves eliciting reasoning and reflection guided by safety rules, then leveraging fine-tuning to enhance safety reasoning. Repeating this process creates a synergistic cycle. Improvements in the model's reasoning and interpretation of safety rules allow it to produce better reasoning data under safety rule prompts, which is then utilized for further training. Experiments show that STAR-S effectively defends against jailbreak attacks, outperforming baselines. Code is available at: https://github.com/pikepokenew/STAR_S.git.

CARE-Bench: A Benchmark of Diverse Client Simulations Guided by Expert Principles for Evaluating LLMs in Psychological Counseling

The mismatch between the growing demand for psychological counseling and the limited availability of services has motivated research into the application of Large Language Models (LLMs) in this domain. Consequently, there is a need for a robust and unified benchmark to assess the counseling competence of various LLMs. Existing works, however, are limited by unprofessional client simulation, static question-and-answer evaluation formats, and unidimensional metrics. These limitations hinder their effectiveness in assessing a model's comprehensive ability to handle diverse and complex clients. To address this gap, we introduce \textbf{CARE-Bench}, a dynamic and interactive automated benchmark. It is built upon diverse client profiles derived from real-world counseling cases and simulated according to expert guidelines. CARE-Bench provides a multidimensional performance evaluation grounded in established psychological scales. Using CARE-Bench, we evaluate several general-purpose LLMs and specialized counseling models, revealing their current limitations. In collaboration with psychologists, we conduct a detailed analysis of the reasons for LLMs' failures when interacting with clients of different types, which provides directions for developing more comprehensive, universal, and effective counseling models.

Exploring and Exploiting the Inherent Efficiency within Large Reasoning Models for Self-Guided Efficiency Enhancement

Recent advancements in large reasoning models (LRMs) have significantly enhanced language models' capabilities in complex problem-solving by emulating human-like deliberative thinking. However, these models often exhibit overthinking (i.e., the generation of unnecessarily verbose and redundant content), which hinders efficiency and inflates inference cost. In this work, we explore the representational and behavioral origins of this inefficiency, revealing that LRMs inherently possess the capacity for more concise reasoning. Empirical analyses show that correct reasoning paths vary significantly in length, and the shortest correct responses often suffice, indicating untapped efficiency potential. Exploiting these findings, we propose two lightweight methods to enhance LRM efficiency. First, we introduce Efficiency Steering, a training-free activation steering technique that modulates reasoning behavior via a single direction in the model's representation space. Second, we develop Self-Rewarded Efficiency RL, a reinforcement learning framework that dynamically balances task accuracy and brevity by rewarding concise correct solutions. Extensive experiments on seven LRM backbones across multiple mathematical reasoning benchmarks demonstrate that our methods significantly reduce reasoning length while preserving or improving task performance. Our results highlight that reasoning efficiency can be improved by leveraging and guiding the intrinsic capabilities of existing models in a self-guided manner.

Psychological Counseling Cannot Be Achieved Overnight: Automated Psychological Counseling Through Multi-Session Conversations

In recent years, Large Language Models (LLMs) have made significant progress in automated psychological counseling. However, current research focuses on single-session counseling, which doesn't represent real-world scenarios. In practice, psychological counseling is a process, not a one-time event, requiring sustained, multi-session engagement to progressively address clients' issues. To overcome this limitation, we introduce a dataset for Multi-Session Psychological Counseling Conversation Dataset (MusPsy-Dataset). Our MusPsy-Dataset is constructed using real client profiles from publicly available psychological case reports. It captures the dynamic arc of counseling, encompassing multiple progressive counseling conversations from the same client across different sessions. Leveraging our dataset, we also developed our MusPsy-Model, which aims to track client progress and adapt its counseling direction over time. Experiments show that our model performs better than baseline models across multiple sessions.

How Does Sequence Modeling Architecture Influence Base Capabilities of Pre-trained Language Models? Exploring Key Architecture Design Principles to Avoid Base Capabilities Degradation

Pre-trained language models represented by the Transformer have been proven to possess strong base capabilities, and the representative self-attention mechanism in the Transformer has become a classic in sequence modeling architectures. Different from the work of proposing sequence modeling architecture to improve the efficiency of attention mechanism, this work focuses on the impact of sequence modeling architectures on base capabilities. Specifically, our concern is: How exactly do sequence modeling architectures affect the base capabilities of pre-trained language models? In this work, we first point out that the mixed domain pre-training setting commonly adopted in existing architecture design works fails to adequately reveal the differences in base capabilities among various architectures. To address this, we propose a limited domain pre-training setting with out-of-distribution testing, which successfully uncovers significant differences in base capabilities among architectures at an early stage. Next, we analyze the base capabilities of stateful sequence modeling architectures, and find that they exhibit significant degradation in base capabilities compared to the Transformer. Then, through a series of architecture component analysis, we summarize a key architecture design principle: A sequence modeling architecture need possess full-sequence arbitrary selection capability to avoid degradation in base capabilities. Finally, we empirically validate this principle using an extremely simple Top-1 element selection architecture and further generalize it to a more practical Top-1 chunk selection architecture. Experimental results demonstrate our proposed sequence modeling architecture design principle and suggest that our work can serve as a valuable reference for future architecture improvements and novel designs.

MPO: Multilingual Safety Alignment via Reward Gap Optimization

Large language models (LLMs) have become increasingly central to AI applications worldwide, necessitating robust multilingual safety alignment to ensure secure deployment across diverse linguistic contexts. Existing preference learning methods for safety alignment, such as RLHF and DPO, are primarily monolingual and struggle with noisy multilingual data. To address these limitations, we introduce Multilingual reward gaP Optimization (MPO), a novel approach that leverages the well-aligned safety capabilities of the dominant language (English) to improve safety alignment across multiple languages. MPO directly minimizes the reward gap difference between the dominant language and target languages, effectively transferring safety capabilities while preserving the original strengths of the dominant language. Extensive experiments on three LLMs, LLaMA-3.1, Gemma-2 and Qwen2.5, validate MPO's efficacy in multilingual safety alignment without degrading general multilingual utility.