The paper introduces the Modality Discrepancy Transformer (MDT), a model designed to detect ambivalence and hesitancy in clinical videos by capturing cross‑modal disagreement across facial, vocal, and linguistic signals. MDT expands a 6‑token representation to 9 tokens that include modality embeddings, absolute‑difference features, and Hadamard‑product discrepancy features, which are processed through Transformer self‑attention with FiLM‑based text conditioning and LoRA fine‑tuning. On the BAH dataset from the 3rd ABAW Challenge, MDT achieves a Macro F1 score of 0.7408 on the labelled test split and 0.7368 on the private leaderboard, surpassing the strongest baseline by over 10 points while training in under 20 minutes on a single GPU.
By Shiyu Luo, Yu Wang, Jiawen Huang, Zhaoxiang Xiao, Chenxi Huang, Qi Zhang, Bin Liu
Neo-Classic is a new benchmark designed to evaluate linguistic‑aesthetic reasoning in Classical Chinese poetry. It uses an out‑of‑sample dataset of strictly metrical poems written by contemporary experts and a set of reverse‑understanding probes, avoiding reliance on historical corpora. Experiments with leading LLMs show a 20–50% performance drop on contemporary texts and low accuracy (0–13%) on discourse‑level ordering, indicating that current models excel at local patterns but struggle with global hierarchical planning.
By Han Zhang, Zihan Gu, Zhiyuan Wang, Tianyi Ma, Jiacheng Lu, Xinyan Zhang, Yuhao Wei, Cheng Hua
The paper introduces a frontend‑backend architecture for full‑duplex speech‑to‑speech models that enables tool calls while preserving natural conversational flow. The frontend emits a delegation token and streams ASR transcripts to a text‑based backend LLM, which performs tool calls and returns results that are re‑injected into the frontend via a lightweight prefill‑and‑repeat mechanism before streaming TTS synthesis. In single‑turn evaluations the system achieves 92‑97% tool‑call recall, 81.2% accuracy in rejecting irrelevant calls, and competitive performance on Full‑Duplex‑Bench‑V3 and EVA‑Bench when paired with a large backend model.
By Ke Hu, Slyne Deng, Chen Chen, Elena Rastorgueva, Edresson Casanova, Punit Kumar, Dharmendra Choudhary, Nikhil Srihari, Ameya Sunil Mahabaleshwarkar, Viet Anh Trinh, Slim Essid, Oluwatobi Olabiyi, Zhehuai Chen
Reflective Recovery is a self‑supervised method that turns failed reasoning attempts into training data, enabling large language models to learn how to correct mistakes during inference. By extracting initial segments of erroneous trajectories and using them as prompts, the approach teaches models to recognize and recover from errors without external critics. Experiments show significant accuracy gains on benchmarks such as AIME 2025 and Minerva, and the method overcomes the scaling collapse problem, fostering emergent self‑correction behaviors.
By Qirui Chen, Renjie Pi, Jiahui Gao, Lingpeng Kong
The study investigates whether pretrained transformer models encode functional words—such as pronouns and adverbs—in a way that mirrors human usage. By comparing embeddings of nouns with those of their functional counterparts in both isolated and parallel sentences, the authors find that functional words occupy a central yet distinct position in embedding space and that parallel lexicalized and functional sentences reside in different subspaces. Experiments show that only a mixed training set of functional and lexicalized sentences reveals shared syntactic and semantic structure, whereas training on either type alone fails to capture this parallelism.
By Giuseppe Samo, Vivi Nastase, Paola Merlo
Compass‑v3 is a 245B‑parameter Mixture‑of‑Experts language model tailored for Southeast Asian e‑commerce, featuring 71B active parameters per token and hardware‑efficient expert parallelism. It is trained on 12 trillion multilingual tokens and synthetic e‑commerce instructions, and incorporates Optimal‑Transport Direct Preference Optimization to improve instruction adherence. Benchmarks show it outperforms GPT‑4, DeepSeek‑V3.1, and Qwen3‑235B, and it is already deployed at scale on Shopee, handling over 70% of the platform’s LLM traffic.
By Sophia Maria
CORTEX is a novel framework that transforms web‑scale corpus construction from flat document filtering into structured knowledge organization using an Ontological Corpus Graph (OCG). The OCG comprises a quality‑refined content layer, a lightweight ontology layer that evolves via LLMs, and a cross‑domain alignment layer that supports arbitrary taxonomic resolution. Experiments demonstrate CORTEX’s effectiveness, and the authors release a 24.14 B‑token refined corpus, its OCG, and a cross‑domain benchmark called CortexBench for evaluating large language models.
By Chengtao Gan, Xiaoke Guo, Yushan Zhu, Zhaoyan Gong, Zhiqiang Liu, Songze Li, Huajun Chen, Wen Zhang
The paper introduces an adaptive region‑dividing strategy that projects a 3D point cloud onto a bird’s‑eye‑view plane to detect building regions, then back‑projects bounding boxes to create structure‑aligned training blocks for unified scene‑level evaluation. It also proposes a fine‑grained classification model using a point transformer classifier and a spatially‑supervised contrastive loss to improve inter‑class discriminability, addressing class imbalance with a weighted cross‑entropy. Experiments on UrbanBIS and STPLS3D datasets show the method outperforms state‑of‑the‑art approaches in both building instance segmentation and fine‑grained classification.
By Weiyuan Zhang, Qi Zhang, Hui Huang
The paper introduces a new task called visual scene negative captioning, which aims to describe what should be present in an image but is actually absent, a capability crucial for safety-critical applications. It proposes the CRCD framework, which uses counterfactual reconstruction and contrastive decoding to overcome affirmation bias, limited mental filling, and representation bias. CRCD employs a dual-branch architecture for amodal completion and functional association, along with multi-condition representation learning, to generate accurate negative captions and sets a high-performance baseline for this emerging task.
By Zhiyun Jiang, Hanyong Wang, Binbin Liang, Yu Xie, Menglong Yang, Wei Li
QCPruner is a training‑free visual token pruning method that conditions both token selection and representation on the query via bilateral utility weighting. It fuses keyword‑matched query anchors with cross‑modal cues to compute a nonnegative facility‑location objective that is monotone and submodular, guaranteeing a (1‑1/e) greedy approximation. Across multiple multimodal large language models, QCPruner consistently outperforms existing pruning methods, achieving over 96% of unpruned performance even with very few tokens retained.
By Shengli He (Guizhou University), Yongchao Liang (Guizhou University), Roumeng He (Shanghai Ocean University), Junjie Zeng (Guizhou University), Jiyuan He (Guizhou University), Can Wu (Guizhou University), Li Zheng (Guizhou University)
The paper introduces Latent-Centroid Steering (LCS), a single-pass classifier-free guidance method for vision‑language autonomous driving models. LCS replaces instance‑level residuals with class‑level latent shifts, projecting conditional representations toward precomputed command‑specific centroids to enhance command adherence. Experiments on Bench2Drive and nuScenes show that LCS cuts inference latency by about 50% while improving driving performance.
By Meibo Hu, Jiamian Wang, Pichao Wang, Zhiqiang Tao
The paper investigates whether large language models (LLMs) can internally detect harmful content, bypassing external guardrails that add latency and computational cost. By extracting activations from LLaMA‑3.1‑8B and training lightweight MLP probes, the authors achieve high F1 scores (99%, 83%, and 84%) on WildJailbreak, Beavertails, and AEGIS 2.0 benchmarks, rivaling much larger guard models while reducing overhead. This suggests that internal state monitoring can provide efficient safety checks for resource‑constrained, time‑critical deployments.
By Alizishaan Khatri, Chiquita Prabhu, Omkar Neogi
FINSKILLOPS is a multi‑agent system designed to improve SEC filing question‑answering after deployment by creating reusable skills from evidence‑grounded failure diagnoses. It manages these skills through targeted validation, regression checks, negative controls, and versioned replacement or retirement, ensuring that new patches do not introduce regressions. Across six benchmarks, a frozen skill registry outperforms other systems, and in a 12‑round operational study the system reduced the non‑correct rate from 20.0% to 12.5% while promoting only six of 33 proposed skills.
By Yanzhang Ma, Zhenghan Tai, Hanwei Wu, Sizhe Guan, Jianliang Lei, Hailin He, Chaolong Jiang, Jijun Chi, Tung Sum Thomas Kwok, Bohuai Xiao, Jingrui Tian, Xinlu Wu, Xingao Zhan, Peng Lu, Muzhi Li, Yihong Wu, Liheng Ma, Sicheng Lyu, Tianshuo Yan, Junhao Zhu, Yaqian Xu, Lei Ding, Yufei Cui, Ziquan Liu, Boyu Han, Hengli Liu, Ling Zhou, Xinyu Wang
The paper introduces the Generic Multi-Agent Trading System (GMATS), a framework for studying how large language model (LLM) based trading stacks react to black-box, input-only attacks that inject plausible social‑media content. It defines contagion metrics—belief‑shift scores at analyst and coordinator layers and attack‑clean deltas on backtest metrics—to trace the spread of adversarial signals. Experiments on a safe offline benchmark show that even simple attackers can significantly degrade risk‑return profiles, while certain multi‑agent topologies and coordinator prompts can mitigate these effects.
By Qi Rong Sua, Junhao Dong, Nguyen Duc Thai, Yuqing Wen, Cheston Tan, Yew-Soon Ong
The paper investigates how large‑language‑model (LLM) based AI agents mix latency, local resource usage, and container bottlenecks when processing user requests that involve remote LLM calls and local tool execution. By measuring three representative tasks—retrieval‑augmented question answering, web search, and software coding—the authors show that agents exhibit diverse resource dynamics, with concurrent requests revealing task‑specific bottlenecks in CPU, disk I/O, and memory. Leveraging these insights, they propose CPU‑aware tool admission and task‑aware CPU allocation, achieving up to a 5.4× speed‑up for CPU‑sensitive tasks and a 32% reduction in average latency across multiple tasks.
By Wonmi Choi, Minuk Park, Zhixiong Niu, Yongqiang Xiong, Chuck Yoo, Gyeongsik Yang
Code-as-Auditor is an LLM-based framework that transforms regulatory information into formal checklists and executable decision trees, encoding rules as interpretable code. During inference, the model expands each checklist item into factual and counterfactual questions, guiding reasoning over case-specific evidence and potential violations. This pipeline moves from evidence identification to rule application and final decision-making, with a self‑verification loop that enhances logical consistency and traceability, leading to more accurate and evidence‑backed compliance evaluations in privacy and data protection scenarios.
By Jisoo Kim, Taeyoon Kwack, Jinwoo Jang, Woo Kyung Kim, Honguk Woo
The paper investigates what gradient similarity measures in data attribution for large language models. By independently varying task and answer format in supervised fine‑tuning benchmarks, the authors show that gradient alignment is driven by answer format rather than task semantics, with strong alignment for shared formats and none for differing formats. This pattern persists across training stages, model sizes, and families, and is evident in the selections of the LESS data‑selection method, which over‑represents its own answer format.
By Sunwoo Kim, Seokwon Jung, Sohyung Kim, Seong Joon Oh, Alice Oh
The paper introduces Live-ProsodyJudge (LPJ), a cost‑effective pairwise evaluator distilled from Gemini for assessing fine‑grained prosody in live streaming TTS. It identifies a flaw called verdict coupling, where multi‑dimensional scores collapse into a single preference, and proposes Decoupled‑Live‑ProsodyJudge (D‑LPJ) to eliminate this issue through masking and a span‑local GRPO strategy. Experiments show LPJ outperforms a single Gemini call in accuracy, and D‑LPJ provides independent dimension judgments, achieving high alignment with human top‑3 selections in a TTS candidate tournament.
By Zifan Guan, Longyu Lu, Junan Zhang, Zhizheng Wu, Meiguang Jin, Junfeng Ma
The paper introduces InterTrack, a behavior world model that enables humanoid robots to perform robust whole-body tracking while interacting with varied terrain and objects. Using a Transformer architecture, InterTrack predicts actions, states, and behavior distributions conditioned on the environment, and it is trained with an automated pipeline that reconstructs 3D support geometry from retargeted motions. The system achieves an 81.3% success rate on terrain interaction, a 99.3% fall-recovery rate, and outperforms leading baselines in both free-space tracking and cross-terrain scenarios.
By Ziyang Cheng, Tianshu Tang, Jinxin Lan, Xinze Chen, Yuhan Gong, Zhichao Liu, Changzhong Wu, Yahao Mao, Zongyan Deng, Mingxuan Ma, Huasen Xi, Yilong Liu, Yutong Wu, Xiaofeng Wang, Borui Zhang, Bingyao Yu, Yang Wang, Yun Ye, Guan Huang, Xiaojie Jin, Zheng Zhu, Jiwen Lu
SETTer is a transformer-based model designed for long‑term multivariate time‑series forecasting. It introduces decoupled self‑attention and hybrid masking to better handle high dimensionality and complex relationships, while adding explainable structures to highlight discriminative patterns. Experiments on real‑world benchmarks show that SETTer outperforms state‑of‑the‑art models in 88% of scenarios.
By Abraham Ezema, Chijioke Eze, Ferdinanda Ponci, Antonello Monti