In large language model (LLM) agents, reasoning trajectories are treated as reliable internal beliefs for guiding actions and updating memory.
In this paper, inspired by the vulnerability of unfaithful intermediate reasoning trajectories, we propose Self-Audited Verified Reasoning (SAVeR), a novel framework that enforces verification over internal belief states within the agent…
However, existing benchmarks remain constrained to isolated scenarios, narrow action spaces, or synthetic data, failing to capture the holistic nature of authentic human behavior.
To bridge this gap, we introduce OmniBehavior, the first user simulation benchmark constructed entirely from real-world data, integrating long-horizon, cross-scenario, and heterogeneous behavioral patterns into a unified framework.
Pairwise PreferenceLlm As JudgeAutomatic MetricsGeneral
Long-term memory is essential for conversational agents to maintain coherence, track persistent tasks, and provide personalized interactions across extended dialogues.
Experiments on the LoCoMo benchmark show that HyperMem achieves state-of-the-art performance with 92.73% LLM-as-a-judge accuracy, demonstrating the effectiveness of HyperMem for long-term conversations.
Xuan Feng, Shuai Zhao, Luwei Xiao, Tianlong Gu, Bo An · Apr 9, 2026 · Citations: 0
Pairwise PreferenceGeneral
Unlike standard preference optimization which applies broad penalties, Self-Debias employs a fine-grained trajectory-level objective subject to dynamic debiasing constraints.
Zhifei Xie, Zongzheng Hu, Fangda Ye, Xin Zhang, Haobo Chai, Zihang Liu · Apr 9, 2026 · Citations: 0
Automatic MetricsGeneral
Prior work remains largely confined to laboratory settings, leaving a clear gap in real-world proactive agent: depth, complexity, ambiguity, precision and real-time constraints.
We first propose DD-MM-PAS (Demand Detection, Memory Modeling, Proactive Agent System) as a general paradigm for streaming proactive AI agent.
Xinliang Frederick Zhang, Lu Wang · Apr 9, 2026 · Citations: 0
Pairwise PreferenceGeneral
Personalized large language models (PLLMs) have garnered significant attention for their ability to align outputs with individual's needs and preferences.
Extensive evaluations on long-horizon benchmarks using the Qwen-3 model family (4B to 32B) validate the effectiveness of TSUBASA, surpassing competitive memory-augmented systems that rely primarily on memory writing, such as Mem0 and…
Pluralistic alignment has emerged as a critical frontier in the development of Large Language Models (LLMs), with reward models (RMs) serving as a central mechanism for capturing diverse human values.
To bridge this gap, we introduce Personalized RewardBench, a novel benchmark designed to rigorously assess reward models' capacity to model personalized preferences.
We introduce a multi-category, cross-cultural benchmark for this task and evaluate VLMs using an LLM-as-Judge framework that measures semantic alignment with reference annotations.
Recent LLM-based agents often place world modeling, planning, and reflection inside a single language model loop.
We introduce a declared reflective runtime protocol that externalizes agent state, confidence signals, guarded actions, and hypothetical transitions into inspectable runtime structure.
As large language models (LLMs) evolve from static chatbots into autonomous agents, the primary vulnerability surface shifts from final outputs to intermediate execution traces.
To address this gap, we introduce TraceSafe-Bench, the first comprehensive benchmark specifically designed to assess mid-trajectory safety.
Xuanbo Su, Wenhao Hu, Le Zhan, Yuting Xie, Kailin Lyu, Kaijie Chen · Apr 8, 2026 · Citations: 0
Human EvalSimulation EnvGeneral
We introduce SalesLLM benchmark, a bilingual (ZH/EN) benchmark derived from realistic applications covering Financial Services and Consumer Goods, built from 30,074 scripted configurations and 1,805 curated multi-turn scenarios with…
We propose a fully automatic evaluation pipeline that combines (i) an LLM-based rater for sales-process progress,and (ii) fine-tuned BERT classifiers for end-of-dialogue buying intent.
Specifically, DRCR employs two complementary feedback signals, discourse coherence and response quality, to construct preference data for both context rewriting and response generation.
Shicheng Liu, Yucheng Jiang, Sajid Farook, Camila Nicollier Sanchez, David Fernando Castro Pena, Monica S. Lam · Apr 7, 2026 · Citations: 0
Human EvalGeneral
Deep research with Large Language Model (LLM) agents is emerging as a powerful paradigm for multi-step information discovery, synthesis, and analysis.
In this paper, we present DataSTORM, an LLM-based agentic system capable of autonomously conducting research across both large-scale structured databases and internet sources.
Danqing Wang, Da Yin, Ruta Desai, Lei Li, Asli Celikyilmaz, Ansong Ni · Apr 7, 2026 · Citations: 0
Automatic MetricsGeneral
Motivated by this, we propose an interruptible communication framework that allows the agent who is listening to interrupt the current speaker.
We evaluate our framework across various multi-agent scenarios, including 2-agent text pictionary games, 3-agent meeting scheduling, and 3-agent debate.
Arabic-DeepSeek-R1 achieves the highest average score across the seven-benchmark OALL suite while establishing SOTA or near-SOTA, including dominant results on grammar-focused MadinahQA (surpassing both GPT-5.1 and the OALL leader by…
Our results indicate that the combination of sparse MoE architecture, culturally-informed CoT distillation with explicit Arabic linguistic checks, and strategic bilingual data curation enables an open-source adapted model to systematically…
These models remarkably achieve high enough accuracy on the Cross-Domain Pedagogical Knowledge (CDPK) Benchmark to establish new state-of-the-art (SOTA) results across the interactive Pedagogy Benchmark Leaderboard and surpass significantly…
