Mem++ is a non‑destructive memory framework for large language model agents that records every document in full, along with its date and author, instead of compressing it at write time. At read time, it retrieves only documents up to the requested time and fuses lexical and semantic rankings, leaving the choice of which version to use to the answering model. Evaluations on OrgMemBench show Mem++ outperforms the strongest baseline by 8.0 to 13.1 points and achieves the best overall score with gpt‑4.1‑mini, while also ranking highly on other benchmarks such as LoCoMo and LongMemEval‑S.
By Ahmad Yehia, Aly O. Abdelkareem, Islam Ahmed, Hesham Omran, Khaled Alashmouny, Christian Claudel, Abduallah Mohamed
The paper introduces Graph of Concept Predictors (GCP), a reasoning-aware active distillation framework that captures a large language model’s intermediate reasoning as a directed acyclic graph of concepts and mirrors this structure in a smaller student model. GCP improves sample efficiency by using a graph-aware acquisition strategy that weighs concept uncertainty, gradient diversity, and node centrality, and enhances training stability through targeted sub‑module retraining that updates only the most influential concept predictors. Experiments on eight NLP classification benchmarks show that GCP achieves better performance under limited annotation budgets while providing more interpretable and controllable training dynamics.
By Ziyang Yu, Liang Zhao
The study examines whether asking large language models (LLMs) to predict whether readers would share misinformation adds useful information beyond predicting credibility. Eight LLM versions were audited on 290 synthetic misinformation articles, yielding 1,256 paired survey responses. While credibility scores explained a substantial portion of sharing variation, adding sharing predictions to credibility did not meaningfully reduce error, indicating that sharing answers contain structured variation but lack established incremental validity for human sharing.
By Zonghuan Xu, Xiang Zheng, Yutao Wu, Xingjun Ma
ESCROW is a post‑deployment maintenance framework for large‑language‑model agents that operate under policy‑bound enterprise workflows. It ensures that any skill revisions proposed by the LLM are only deployed after empirical evaluation, using distributed diagnosis, consensus, per‑category non‑regression guards, cross‑cycle anti‑regression checks, and an accuracy–cost Pareto search to produce a versioned, auditable diff for each change. In a real production financial document‑auditing system, ESCROW achieved the best accuracy–cost trade‑off compared with baseline methods and demonstrated transferability on a public $ au$‑bench.
By Ruoqi Shu, Chen Dan, Xuhui Wang, Tianhua Xu, Mengxi Luo, Yanming Mai, Bo Wan
The paper introduces PrecepTron, a 32‑billion‑parameter language model fine‑tuned with low‑rank adaptation to evaluate clinical reasoning in large language models (LLMs) at a physician level. It also releases GRAND‑ROUNDS, a benchmark of 9,217 scored responses from 160 clinicians across seven studies. Using PrecepTron, the authors replicate key findings from major medical AI studies and explore new questions about LLM diagnostic accuracy, demonstrating that fine‑tuned models can provide consistent, scalable physician‑level scoring.
By Thomas A. Buckley, Zahir Kanjee, Peter G. Brodeur, Byron Crowe, Anthony M. Pettinato, Aashna P. Shah, Adrian D. Haimovich, Liam G. McCoy, Daniel Restrepo, Jason A. Freed, Ethan Goh, Jonathan H. Chen, Laura Zwaan, Katherine E. Goodman, Daniel J. Morgan, Raja-Elie E. Abdulnour, Adam Rodman, Arjun K. Manrai
The paper investigates the gap between intent and behavior in jailbreak attacks on LLM-based robots, showing that many intent jailbreaks fail to produce harmful physical actions due to robot-specific constraints. It introduces POEF, an automated red‑teaming framework that optimizes jailbreak prompts using hidden‑layer gradients and evaluates policy feasibility with a multi‑agent system, achieving an 80% success rate on commercial robots. The authors also propose defense strategies and emphasize the need for stronger countermeasures before widespread deployment of LLM-based robots.
By Xuancun Lu, Zhengxian Huang, Xinfeng Li, Chi Zhang, Xiaoyu Ji, Wenyuan Xu
The paper introduces MolPAIR, a framework that enhances tabular foundation models (TFMs) for predicting molecular properties under structural shifts by incorporating molecule-level and molecular-pair contexts without task‑specific parameter updates. MolPAIR refines initial predictions from a global TFM by using a second frozen TFM to predict differences in prediction errors between a query molecule and labeled reference molecules. Experiments on 58 MoleculeACE and Polaris tasks show that MolPAIR improves performance on 46 tasks across multiple molecular representations and TFM backbones, demonstrating that explicit molecular comparisons strengthen tabular in‑context learning for structural generalization.
By Jinmo Lee, Dooho Lee, Minho Jeong, Jaemin Yoo
The paper introduces Importance-Aware Class-Balanced Sparsification (ICS) for wireless split learning, a lightweight method that ranks feature channels on the server using Grad‑CAM scores from the true‑class logit. By aggregating per‑class scores into a class‑balanced importance vector, clients can retain the top‑N feature channels without extra forward or backward passes, reducing communication overhead while mitigating head‑class bias under label skew. The authors provide a non‑asymptotic convergence bound, analyze overhead against baselines, and extend the approach to parallel split learning and transformer models, demonstrating consistent performance gains, especially under severe non‑i.i.d. data partitions.
By Bumjun Kim, Yoon Huh, Wan Choi
The study investigates whether a second review by a different large language model (LLM) improves error detection in code, documentation, and analyses. Using 30 artifacts with 150 planted errors and 900 review sessions across three reviewer models, the authors find that a top-tier cross-model reviewer performs similarly to same-model reviews in fresh sessions, identifies partly different errors, and that a combination of a fresh same-model review and a cross-model review detects more errors than two fresh same-model reviews. However, the results do not conclusively separate the effects of model difference from reviewer capability, and a lightweight cross-model reviewer does not outperform same-model review.
whyItMatters":"The findings suggest that while cross-model reviews can enhance error detection, the benefit depends on reviewer skill and the specific review strategy, informing how teams might structure LLM-based code review processes."
By Tae-Eun Song
The paper investigates demographic bias in large language model (LLM) predictions of ICU mortality and introduces Case-Based Prompting (CAP), a training‑free method that adds clinically similar historical cases as contextual evidence. CAP improves predictive performance on the MIMIC‑IV dataset, raising AUROC from 0.806 to 0.873 and AUPRC from 0.497 to 0.694, while reducing subgroup disparities for sex and White‑Black comparisons. The study also shows that simply removing demographic information does not uniformly enhance fairness or accuracy, and that feature‑dependence patterns remain consistent across groups.
By Gangxiong Zhang, Yongchao Long, Yuxi Zhou, Yong Zhang, Shenda Hong
MosaiChunk introduces a spatio-temporal memory mechanism that composes a mosaic of selected historical key-value entries across space and time to improve long-horizon autoregressive video generation. By keeping the video generator frozen and learning a lightweight router to select historical sections within a fixed memory budget, the method enables the generator to recover fine-grained visual details when objects or scenes reappear. Experiments on the RememBench benchmark demonstrate that MosaiChunk consistently outperforms sliding-window inference and whole-chunk retrieval in both text-to-video and image-to-video settings.
By Yiwen Zhang, Haocheng Xi, Michael Tian-Yue Liu, Alexei A. Efros, Hadar Averbuch-Elor, Qianqian Wang, Haiwen Feng
LangMap introduces a human‑verified benchmark for language‑conditioned goal navigation (LGN) that spans four hierarchical semantic levels—scene, room, region, and instance—within real‑world indoor 3D scans. The dataset, built on HM3D, contains 18K tasks with concise and detailed descriptions for 414 object categories, and its contrastive annotation protocol ensures high‑quality, discriminative region and instance labels. Evaluation shows that LangMap’s descriptions improve text‑to‑view matching accuracy by 23 points over GOAT‑Bench and achieve a 92.5% unique‑and‑correct match rate in an independent human audit, while a proposed RGB‑only baseline, PlaNaVid, attains top‑tier success rates without depth or 3D scene representations.
By Bo Miao, Weijia Liu, Jun Luo, Lachlan Shinnick, Jian Liu, Thomas Hamilton-Smith, Yuhe Yang, Zijie Wu, Vanja Videnovic, Feras Dayoub, Anton van den Hengel
The paper introduces a 3D keypoint detector that replaces traditional heuristic post‑processing with a learned suppression module. This module, implemented as a graph neural network over candidate keypoints, learns to keep, suppress, or relocate points, and is built on a Point Transformer backbone extended with a directional graph neural network. Experiments show the approach outperforms DBSCAN and greedy non‑maximum suppression, surpasses KeypointDETR on most KeypointNet categories, and achieves top performance on the Building3D benchmark while remaining competitive on the Tallinn split.
By Batuhan Arda Bekar, Can Sar{\i}, H\"useyin Can G\"ulkan, Bar{\i}\c{s} \"Ozcan
The paper introduces REHAB26-ViewAngles, a dataset of correct and incorrect rehabilitation exercise videos captured from many camera angles, and a new separability metric to evaluate pose‑estimation algorithms. It evaluates single‑camera 2D and 3D pose estimation and four multi‑camera approaches, showing that an optimally placed 2D camera can boost separability by 16.9% over a frontal view and that combining two views can improve accuracy by up to 13.1%. These findings provide concrete guidance for setting up camera systems in home and clinical rehabilitation monitoring.
By Miriama J\'ano\v{s}ov\'a, Andreas Lang, Petra Budikova, Jan Sedmidubsky
The paper introduces AEGIS, a buffer‑free, layer‑wise orthogonal gradient projection framework that enables continuous fine‑tuning of Vision‑Language Models for robotic manipulation while protecting pre‑trained visual reasoning. AEGIS uses pre‑training activation statistics as static anchors and applies a Wasserstein‑2 transport penalty to generate an anchor‑restoration gradient, followed by a dual‑backward pass that orthogonalizes task gradients against this restoration vector. Experiments on the LIBERO manipulation benchmark with PaliGemma2‑3B show that AEGIS preserves Visual Question Answering performance and baseline loss while achieving continuous action convergence, all without replay buffers, teacher models, or co‑training data.
By Guransh Singh
The paper introduces memetic trojans, a new class of attacks that embed adversarial payloads within socially contagious content shared by large language model agents. Unlike traditional agent worms that rely on malicious retransmission, memetic trojans exploit agents’ natural tendency to share content that aligns with their internal motivations, leading to high virality and significant amplification of exposure. Experiments on the Moltbook platform show that such contagions can be retransmitted in about 50% of subsequent posts and upvoted 2.5 times more than average, with simulations indicating up to a 3.19x increase in expected exposure across the network.
whyItMatters":"The study reveals that endogenous social transmission is a distinct security vulnerability in multi‑agent systems, implying that defenses must address network-level dynamics rather than just prompt‑injection detection or agent compromise prevention."
By Birk Torpmann-Hagen, Finn Schwall, Leon Moonen
The paper "Where's Waldo? Query-language Preference under Cross-lingual Knowledge Disparities" introduces the Waldo benchmark, a multilingual QA dataset built from Wikipedia that focuses on knowledge gaps and conflicts across languages. It evaluates eight models in five languages and finds that when a fact is missing in one language, models tend to use evidence from the other language, but when conflicting accounts exist, responses align strongly with the query language, leading to different answers for semantically identical questions. The study also explores mitigation strategies, including ablating attention heads and LoRA-based training, which can reduce the preference gap by up to 61.5%.
By Dayeon Ki, Ruochen Zhang, Silviu Cucerzan, Ryen W. White, Ning Gao
The paper presents a matched‑budget audit framework for evaluating recaptioned image‑text supervision distributions, which are created by applying a captioning policy, captioner, and source corpus to generate dense descriptions for text‑to‑image training. The framework operates under a fixed text budget of 64 tokens and produces a five‑axis profile covering prompt coverage, image‑conditioned faithfulness, and caption surface quality, using controllable basic units as a common claim metric. The authors apply the framework to seven paired comparisons across five public corpora, demonstrating improvements in supported claim units and revealing consistent long‑vs‑dense trade‑offs across different judges and budgets, and they release the audited multi‑source recap corpus and audit artifacts.
By Giyeong Oh, Junghun Park, Yuhan Bae, Youngjae Yu
The paper investigates how the choice of reconstruction loss, specifically mean squared error (MSE), affects optimization in sequential post‑training quantization (PTQ) of large language models (LLMs). It identifies an Optimization Imbalance where MSE causes reconstruction loss magnitudes—and thus gradient magnitudes—to vary dramatically across quantization stages, leading to uneven parameter updates. The authors propose that root mean squared error (RMSE) variants, which implicitly normalize gradients, can decouple optimization strength from loss scale and outperform MSE as a drop‑in replacement.
By Chao Li, Shigeng Wang, Anbang Yao
The paper investigates whether reviewers with limited expertise can reliably determine if a coding agent’s patch resolves a bug, using 411 execution-labeled traces from three agents and 101 controlled cases. It finds that providing structured but unchecked evidence can improve defect detection and reduce over-rejection, and that a frozen cascade of static errors and generated tests achieves high coverage and catch rates on held‑out GPT‑5.4 and Gemini traces. Reviewer size alone does not predict quality, and the study highlights the challenge of creating reliable checks without official tests.
By Junyu Guo, Shangding Gu, Ming Jin, Javad Lavaei