Jiawei Feng, Jiancan Wu, Xingyu Zhu, Junkang Wu, Xiang Wang, Xiangnan He · Aug 20, 2026 · Citations: 0
Pairwise PreferenceGeneral
Direct Preference Optimization (DPO) has emerged as an effective approach for aligning large language models (LLMs) with human preferences.
To address these challenges, we propose Perception-Enhanced Alignment DPO (PEA-DPO), a framework for multimodal LLMs alignment, which explicitly leverages visual preference signals to overcome visual insensitivity.
Vulnerability prioritization is inherently preference dependent, since the same CVE can receive different remediation priority under different operational preference scenarios.
In practice, organizations already operate under a preference scenario, but this preference is often implicit and difficult to express as a written prompt instruction, while triage queries usually do not encode it.
Mint-Agent Team, Kun Wang, Gavin Zhang, Yaze Geng, Lei Tang, Yaoyang Yi · Aug 17, 2026 · Citations: 0
Expert VerificationAutomatic MetricsGeneral
We present Mint-Agent, a family of finance-native agentic models designed around these two scales of financial intelligence.
Across professional financial benchmarks, our models demonstrate two defining strengths: (1) Reliability: Mint-Ag achieves 98.33% on RFC-Bench, surpassing GPT-5.6-Sol and Claude-Opus-4.8 by 3.66 and 3.00 points; and (2) Executability:…
Cedar Site Bai, Zhenyu Liao, Duanshun Li, Sheikh Sarwar, Huiyuan Chen, Yuan Chen · Aug 16, 2026 · Citations: 0
Pairwise PreferenceAutomatic MetricsGeneral
However, guiding multi-turn interactions to elicit user preferences effectively remains challenging.
Existing approaches either use separate reinforcement learning agents with templated interactions or optimize for interactivity judged by another LLM, without measuring how much useful information is actually gained.
Jingbo Ji, Lingyi Li, Xilong Cheng, Yuhao Zhou, Wenji Zhang, Yuting Tan · Aug 13, 2026 · Citations: 0
Llm As JudgeAutomatic MetricsGeneral
LLM-based agents increasingly rely on external memory to support long-horizon reasoning and interaction.
Experiments on LoCoMo and LongMemEval-S show that RippleMem achieves the best overall performance across evaluated settings, improving LLM-as-a-Judge accuracy by 3.95% on LoCoMo and up to 11.87% on LongMemEval-S, while reducing graph…
We adapt S2G-RAG's structured sufficiency-and-gap judgment to a frozen Search-R1 pipeline and train a Qwen3.5-2B judge on 3,009 states from 900 disjoint HotpotQA questions.
Thus, the trained S2G-style structured judge reduces retrieval while broadly preserving answer accuracy.
To address this challenge, we propose LigBench, an automated evaluation benchmark that enables fine-grained and reliable evaluation of AI research ideas, consistently applicable across different generation distributions.
In addition, we introduce PAIR-IQ, a dataset tailored for training pairwise idea judgment models and serving as an auxiliary reference to support more objective comparative evaluation.
Long-horizon LLM agents must preserve information from past interactions to support future tasks.
More broadly, our results suggest that the accuracy--cost trade-off of long-term agent memory depends not only on what information is retained, but also on the granularity at which it is consolidated.
Xing Zhang, Yanwei Cui, Guanghui Wang, Peiyang He · Aug 13, 2026 · Citations: 0
Red TeamAutomatic MetricsGeneral
We present a two-tier agentic system that separates a maintained, point-in-time knowledge library from report writing.
A portable multi-agent "writer" runtime then composes a contradiction-free, evidence-grounded report at any knowledge cutoff T, reading only evidence with as_of <= T (no look-ahead); red-team verdicts flow back into the librarian.
Yifei Li, Heng Wang, Lingling Zhang, Muye Huang, Xinyu Zhang, Jiashuai Liu · Aug 13, 2026 · Citations: 0
Simulation EnvGeneral
Retrieval can identify a past trajectory that may matter, yet it does not specify how an acting agent should use that trajectory after users, entities, constraints, or environment state have changed.
We identify this post-retrieval reuse step as a distinct bottleneck for long-horizon trajectory memory and formulate an evaluation framework that holds candidate retrieval, target state, model, decoding, and tool budget fixed while varying…
Haolong Chen, Liang Zhang, Zhuo Li, Lei Xue, Guanrxu Zhu · Aug 13, 2026 · Citations: 0
Automatic MetricsGeneral
While Large Language Model (LLM) agents increasingly rely on long-term memory for persistent interactions, the retrieval mechanisms governing this memory are rarely treated as evolvable components.
Notably, it improves the overall average across F1, BLEU-1, and LLM-judge scores by 31.3\% with Qwen3-Next-80B-A3B-Instruct and by 28.1\% with GPT-5.4-nano.
Benchmark contamination is diagnosed today with n-gram overlap, with likelihood-based membership inference, or with canary strings, and each needs something usually unavailable: the training corpus, a well-chosen test statistic, or…
We demonstrate two bottlenecks in existing systems: indices built from context-poor captions are unreliable for agentic search, while retrieval ignores a question's temporal intent.
To address both bottlenecks, we introduce EgoCITE (Egocentric Context-augmented Indexing and Time-aware Evidence retrieval), a long-horizon agentic memory framework for egocentric QA.
To address this, we introduce EpicStar, a framework that enables agents to learn memory as policy to tackle long-horizon reasoning.
Specifically, the agent maintains a bank of successful past episodes as a heuristic alongside a working memory to track short-term environmental changes.
Simon Yu, Nicholas Tomlin, Marwa Abdulhai, Ximing Lu, Derek Chong, Abe Hou · Aug 12, 2026 · Citations: 0
Simulation EnvGeneral
Multi-agent reinforcement learning for human-AI interaction typically relies on a single large language model to simulate user behavior.
Verbalized Sampling improves held-out success by up to 9% over single-simulator RL, and Co-Training pushes gains further to 14%; the human study shows similar gain on real users.
Hung-Chun Hsu, Po-Jen Ko, Che-Cheng Wu, Li-Yang Chang, Chuan-Ju Wang · Aug 12, 2026 · Citations: 0
Llm As JudgeAutomatic MetricsGeneral
Predictive-distribution entropy is a strong answer-selection rule in retrieval-augmented generation (RAG) for question answering: across five QA benchmarks, selecting the answer a frozen respondent LLM produces with the lowest answer-token…
LODESTAR uses reinforcement learning (GRPO) to train, once and offline, a polarizer -- a short fixed natural-language string inserted into the respondent's prompt and never into its weights, directing entropy so that entropy-based answer…
Large language models are increasingly deployed in settings that require simultaneous adherence to multiple explicit constraints - reasoning structure, safety boundaries, output schemas.
We introduce Constraint Saturation Evaluation (CSE), a procedurally generated benchmark that systematically varies the number of simultaneous constraints (k), with every constraint scored by a deterministic, rule-based verifier and zero…
Large language model (LLM) agents integrated with external tools are vulnerable to indirect prompt injections embedded in environmental states.
To bridge this gap, we propose **ToolHazard**, a scalable adversarial environment synthesis framework that reduces human engineering and supports expansion with additional seed domains and compute.
