# Research Papers — 2026-01 · bundle page 109 of 189

> Full-text markdown of papers #5401–5450 of 9403 published in 2026-01, newest first. AI-readable mirrors of the full 9403-paper 2026-01 archive (50 papers per file).

Canonical page: https://aipod.dev/papers/archive/2026-01/page/109
[← Previous bundle](https://aipod.dev/papers/archive/2026-01/page/108.md)
[Next bundle →](https://aipod.dev/papers/archive/2026-01/page/110.md)

### One Instruction Does Not Fit All: How Well Do Embeddings Align Personas and Instructions in Low-Resource Indian Languages?

Authors: Arya Shah, Himanshu beniwal, Mayank Singh
Published: 2026-01-15
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2601.10205v1

Aligning multilingual assistants with culturally grounded user preferences is essential for serving India's linguistically diverse population of over one billion speakers across multiple scripts. However, existing benchmarks either focus on a single language or conflate retrieval with generation, leaving open the question of whether current embedding models can encode persona-instruction compatibility without relying on response synthesis. We present a unified benchmark spanning 12 Indian languages and four evaluation tasks: monolingual and cross-lingual persona-to-instruction retrieval, reverse retrieval from instruction to persona, and binary compatibility classification. Eight multilingual embedding models are evaluated in a frozen-encoder setting with a thin logistic regression head for classification. E5-Large-Instruct achieves the highest Recall@1 of 27.4\% on monolingual retrieval and 20.7\% on cross-lingual transfer, while BGE-M3 leads reverse retrieval at 32.1\% Recall@1. For classification, LaBSE attains 75.3\% AUROC with strong calibration. These findings offer practical guidance for model selection in Indic multilingual retrieval and establish reproducible baselines for future work\footnote{Code, datasets, and models are publicly available at https://github.com/aryashah2k/PI-Indic-Align.

---

### Future-KL Regularized GRPO: Process-Level Credit Assignment from $f$-Divergence Regularization

Authors: Jiarui Yao, Ruida Wang, Hao Bai, Tong Zhang
Published: 2026-01-15
Categories: cs.LG, cs.AI, cs.CL
arXiv: https://arxiv.org/abs/2601.10201v2

Group Relative Policy Optimization (GRPO) is widely used for critic-free Large Language Model (LLM) post-training, but its KL regularization is usually implemented as a local loss-side token penalty. We show that this misses the policy-gradient signal induced by autoregressive KL regularization. Unlike standard KL-regularized Reinforcement Learning (RL) objectives, GRPO's group normalization induces a non-linear prompt-level utility; for binary verifier rewards, this utility is $2\arcsin\sqrt p$. As a result, reward and KL cannot be fused before normalization without changing the implicit objective. We derive the on-policy gradient of GRPO-style objectives with token-wise $f$-divergence regularization. The reward term recovers the standardized GRPO advantage, while the regularizer term includes a causal future-regularization return-to-go omitted by local KL losses. For reverse KL, this yields a simple future KL correction: add a reverse cumulative sum of per-token log ratios after advantage construction. The resulting method, Future-KL Regularized Policy Optimization (FRPO), requires no critic or extra model passes. On mathematical reasoning tasks, FRPO improves pass@16 in our main large-model setting while maintaining higher entropy and lower policy drift than conventional loss-side KL baselines.

---

### ELITE: Efficient Gaussian Head Avatar from a Monocular Video via Learned Initialization and TEst-time Generative Adaptation

Authors: Kim Youwang, Lee Hyoseok, Subin Park, Gerard Pons-Moll, Tae-Hyun Oh
Published: 2026-01-15
Categories: cs.CV
arXiv: https://arxiv.org/abs/2601.10200v1

We introduce ELITE, an Efficient Gaussian head avatar synthesis from a monocular video via Learned Initialization and TEst-time generative adaptation. Prior works rely either on a 3D data prior or a 2D generative prior to compensate for missing visual cues in monocular videos. However, 3D data prior methods often struggle to generalize in-the-wild, while 2D generative prior methods are computationally heavy and prone to identity hallucination. We identify a complementary synergy between these two priors and design an efficient system that achieves high-fidelity animatable avatar synthesis with strong in-the-wild generalization. Specifically, we introduce a feed-forward Mesh2Gaussian Prior Model (MGPM) that enables fast initialization of a Gaussian avatar. To further bridge the domain gap at test time, we design a test-time generative adaptation stage, leveraging both real and synthetic images as supervision. Unlike previous full diffusion denoising strategies that are slow and hallucination-prone, we propose a rendering-guided single-step diffusion enhancer that restores missing visual details, grounded on Gaussian avatar renderings. Our experiments demonstrate that ELITE produces visually superior avatars to prior works, even for challenging expressions, while achieving 60x faster synthesis than the 2D generative prior method.

---

### Graph Regularized PCA

Authors: Antonio Briola, Marwin Schmidt, Fabio Caccioli, Carlos Ros Perez, James Singleton, Christian Michler, Tomaso Aste
Published: 2026-01-15
Categories: cs.LG
arXiv: https://arxiv.org/abs/2601.10199v2

Multivariate data often exhibit complex dependencies that violate the assumption of isotropic residual noise. For such cases, we introduce Graph Regularized PCA (GR-PCA). It is a graph-based regularization of PCA that incorporates the dependency structure of the data features by learning a sparse precision graph and biasing loadings toward the low-frequency Fourier modes of the corresponding graph Laplacian. Consequently, high-frequency signals are suppressed, while graph-coherent low-frequency ones are preserved, yielding interpretable principal components aligned with conditional relationships. We evaluate GR-PCA on synthetic data spanning diverse graph topologies, signal-to-noise ratios, and sparsity levels. Compared to mainstream alternatives, it concentrates variance on the intended support, produces loadings with lower graph-Laplacian energy, and remains competitive in out-of-sample reconstruction. When high-frequency signals are present, the graph Laplacian penalty prevents overfitting, reducing the reconstruction accuracy but improving structural fidelity. The advantage over PCA is most pronounced when high-frequency signals are graph-correlated, whereas PCA remains competitive when such signals are nearly rotationally invariant. The procedure is simple to implement, modular with respect to the precision estimator, and scalable, providing a practical route to structure-aware dimensionality reduction that improves structural fidelity without sacrificing predictive performance.

---

### HumanLLM: Benchmarking and Improving LLM Anthropomorphism via Human Cognitive Patterns

Authors: Xintao Wang, Jian Yang, Weiyuan Li, Rui Xie, Jen-tse Huang, Jun Gao, Shuai Huang, Yueping Kang, Yuanli Gou, Hongwei Feng, Yanghua Xiao
Published: 2026-01-15
Categories: cs.CL
arXiv: https://arxiv.org/abs/2601.10198v4

Large Language Models (LLMs) have demonstrated remarkable capabilities in reasoning and generation, serving as the foundation for advanced persona simulation and Role-Playing Language Agents (RPLAs). However, achieving authentic alignment with human cognitive and behavioral patterns remains a critical challenge for these agents. We present HumanLLM, a framework treating psychological patterns as interacting causal forces. We construct 244 patterns from $\sim$12,000 academic papers and synthesize 11,359 scenarios where 2-5 patterns reinforce, conflict, or modulate each other, with multi-turn conversations expressing inner thoughts, actions, and dialogue. Our dual-level checklists evaluate both individual pattern fidelity and emergent multi-pattern dynamics, achieving strong human alignment ($r=0.90$) while revealing that holistic metrics conflate simulation accuracy with social desirability. HumanLLM-8B outperforms Qwen3-32B on multi-pattern dynamics despite 4$\times$ fewer parameters, demonstrating that authentic anthropomorphism requires cognitive modeling -- simulating not just what humans do, but the psychological processes generating those behaviors. Our dataset, code, and model are available at:https://github.com/YJGoodbye2024/HumanLLM

---

### GFM4GA: Graph Foundation Model for Group Anomaly Detection

Authors: Jiujiu Chen, Weijun Zeng, Shaofeng Hu, Sihong Xie, Hui Xiong
Published: 2026-01-15
Categories: cs.AI
arXiv: https://arxiv.org/abs/2601.10193v1

Group anomaly detection is crucial in many network applications, but faces challenges due to diverse anomaly patterns. Motivated by the success of large language models (LLMs) in natural language processing, graph foundation models (GFMs) is proposed to handle few-shot learning task with fewer labeling efforts. GFMs have been successfully applied to detection of individual anomalies but cannot be generalized to group anomalies, as group anomaly patterns must be detected as a whole and individuals in an abnormal group can look rather normal. Therefore, we propose GFM4GA, a novel graph foundation model for group anomaly detection. The pipeline is pretrained via dual-level contrastive learning based on feature-based estimation and group extraction, to capture potential group anomaly structure and feature inconsistencies. In the downstream tasks, the pipeline is finetuned in parameter-constrained and group-anomaly-proportion weighted few-shot settings, and its adaptive ability to unseen group anomalies expanded via group contexts determined by labeled anomaly neighbors. Experiments show that GFM4GA surpasses group anomaly detectors and GFMs for individual anomalies, achieving average improvements of 2.85% in AUROC and 2.55% in AUPRC.

---

### From Physical Degradation Models to Task-Aware All-in-One Image Restoration

Authors: Hu Gao, Xiaoning Lei, Xichen Xu, Xingjian Wang, Lizhuang Ma
Published: 2026-01-15
Categories: cs.CV
arXiv: https://arxiv.org/abs/2601.10192v1

All-in-one image restoration aims to adaptively handle multiple restoration tasks with a single trained model. Although existing methods achieve promising results by introducing prompt information or leveraging large models, the added learning modules increase system complexity and hinder real-time applicability. In this paper, we adopt a physical degradation modeling perspective and predict a task-aware inverse degradation operator for efficient all-in-one image restoration. The framework consists of two stages. In the first stage, the predicted inverse operator produces an initial restored image together with an uncertainty perception map that highlights regions difficult to reconstruct, ensuring restoration reliability. In the second stage, the restoration is further refined under the guidance of this uncertainty map. The same inverse operator prediction network is used in both stages, with task-aware parameters introduced after operator prediction to adapt to different degradation tasks. Moreover, by accelerating the convolution of the inverse operator, the proposed method achieves efficient all-in-one image restoration. The resulting tightly integrated architecture, termed OPIR, is extensively validated through experiments, demonstrating superior all-in-one restoration performance while remaining highly competitive on task-aligned restoration.

---

### How does downsampling affect needle electromyography signals? A generalisable workflow for understanding downsampling effects on high-frequency time series

Authors: Mathieu Cherpitel, Janne Luijten, Thomas Bäck, Camiel Verhamme, Martijn Tannemaat, Anna V. Kononova
Published: 2026-01-15
Categories: cs.AI
arXiv: https://arxiv.org/abs/2601.10191v2

Automated analysis of needle electromyography (nEMG) signals is emerging as a tool to support the detection of neuromuscular diseases (NMDs), yet the signals' high and heterogeneous sampling rates pose substantial computational challenges for feature-based machine-learning models, particularly for near real-time analysis. Downsampling offers a potential solution, but its impact on diagnostic signal content and classification performance remains insufficiently understood. This study presents a workflow for systematically evaluating information loss caused by downsampling in high-frequency time series. The workflow combines shape-based distortion metrics with classification outcomes from available feature-based machine learning models and feature space analysis to quantify how different downsampling algorithms and factors affect both waveform integrity and predictive performance. We use a three-class NMD classification task to experimentally evaluate the workflow. We demonstrate how the workflow identifies downsampling configurations that preserve diagnostic information while substantially reducing computational load. Analysis of shape-based distortion metrics showed that shape-aware downsampling algorithms outperform standard decimation, as they better preserve peak structure and overall signal morphology. The results provide practical guidance for selecting downsampling configurations that enable near real-time nEMG analysis and highlight a generalisable workflow that can be used to balance data reduction with model performance in other high-frequency time-series applications as well.

---

### HOMURA: Taming the Sand-Glass for Time-Constrained LLM Translation via Reinforcement Learning

Authors: Ziang Cui, Mengran Yu, Chenyu Shi, Yingxuan Shi, Tianjiao Li
Published: 2026-01-15
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2601.10187v3

Large Language Models (LLMs) have achieved remarkable strides in multilingual translation but are hindered by a systemic cross-lingual verbosity bias, rendering them unsuitable for strict time-constrained tasks like subtitling and dubbing. Current prompt-engineering approaches struggle to resolve this conflict between semantic fidelity and rigid temporal feasibility. To bridge this gap, we first introduce Sand-Glass, a benchmark specifically designed to evaluate translation under syllable-level duration constraints. Furthermore, we propose Homura, a reinforcement learning framework that explicitly optimizes the trade-off between semantic preservation and temporal compliance. By employing a constrained reinforcement learning objective featuring a novel dynamic syllable-ratio reward, Homura effectively "tames" the output length. Experimental results demonstrate that Homura significantly outperforms strong baselines, achieving precise length control that respects linguistic density hierarchies without compromising semantic adequacy.

---

### Reinforcement Learning to Discover a North-East Monsoon Index for Rainfall Prediction in Thailand

Authors: Kiattikun Chobtham
Published: 2026-01-15
Categories: cs.LG, astro-ph.EP
arXiv: https://arxiv.org/abs/2601.10181v5

Accurately predicting long-term rainfall is challenging. Global climate indices, such as the El Niño-Southern Oscillation, are standard input features for machine learning. However, a significant gap persists regarding local-scale indices capable of improving predictive accuracy in specific regions of Thailand. This paper introduces a novel North-East monsoon climate index calculated from sea surface temperature to reflect the climatology of the boreal winter monsoon. To optimise the calculated areas used for this index, a Deep Q-Network reinforcement learning agent explores and selects the most effective rectangles based on their correlation with seasonal rainfall. Rainfall stations were classified into 12 distinct clusters to distinguish rainfall patterns between southern and upper Thailand. Experimental results show that incorporating the optimised index into Long Short-Term Memory models significantly improves long-term monthly rainfall prediction skill in most cluster areas. This approach effectively reduces the Root Mean Square Error for 12-month-ahead forecasts.

---

### Bias in the Shadows: Explore Shortcuts in Encrypted Network Traffic Classification

Authors: Chuyi Wang, Xiaohui Xie, Tongze Wang, Yong Cui
Published: 2026-01-15
Categories: cs.LG, cs.NI
arXiv: https://arxiv.org/abs/2601.10180v1

Pre-trained models operating directly on raw bytes have achieved promising performance in encrypted network traffic classification (NTC), but often suffer from shortcut learning-relying on spurious correlations that fail to generalize to real-world data. Existing solutions heavily rely on model-specific interpretation techniques, which lack adaptability and generality across different model architectures and deployment scenarios. In this paper, we propose BiasSeeker, the first semi-automated framework that is both model-agnostic and data-driven for detecting dataset-specific shortcut features in encrypted traffic. By performing statistical correlation analysis directly on raw binary traffic, BiasSeeker identifies spurious or environment-entangled features that may compromise generalization, independent of any classifier. To address the diverse nature of shortcut features, we introduce a systematic categorization and apply category-specific validation strategies that reduce bias while preserving meaningful information. We evaluate BiasSeeker on 19 public datasets across three NTC tasks. By emphasizing context-aware feature selection and dataset-specific diagnosis, BiasSeeker offers a novel perspective for understanding and addressing shortcut learning in encrypted network traffic classification, raising awareness that feature selection should be an intentional and scenario-sensitive step prior to model training.

---

### CC-OR-Net: A Unified Framework for LTV Prediction through Structural Decoupling

Authors: Mingyu Zhao, Haoran Bai, Yu Tian, Bing Zhu, Hengliang Luo
Published: 2026-01-15
Categories: cs.LG
arXiv: https://arxiv.org/abs/2601.10176v2

Customer Lifetime Value (LTV) prediction, a central problem in modern marketing, is characterized by a unique zero-inflated and long-tail data distribution. This distribution presents two fundamental challenges: (1) the vast majority of low-to-medium value users numerically overwhelm the small but critically important segment of high-value "whale" users, and (2) significant value heterogeneity exists even within the low-to-medium value user base. Common approaches either rely on rigid statistical assumptions or attempt to decouple ranking and regression using ordered buckets; however, they often enforce ordinality through loss-based constraints rather than inherent architectural design, failing to balance global accuracy with high-value precision. To address this gap, we propose \textbf{C}onditional \textbf{C}ascaded \textbf{O}rdinal-\textbf{R}esidual Networks \textbf{(CC-OR-Net)}, a novel unified framework that achieves a more robust decoupling through \textbf{structural decomposition}, where ranking is architecturally guaranteed. CC-OR-Net integrates three specialized components: a \textit{structural ordinal decomposition module} for robust ranking, an \textit{intra-bucket residual module} for fine-grained regression, and a \textit{targeted high-value augmentation module} for precision on top-tier users. Evaluated on real-world datasets with over 300M users, CC-OR-Net achieves a superior trade-off across all key business metrics, outperforming state-of-the-art methods in creating a holistic and commercially valuable LTV prediction solution.

---

### ReasAlign: Reasoning Enhanced Safety Alignment against Prompt Injection Attack

Authors: Hao Li, Yankai Yang, G. Edward Suh, Ning Zhang, Chaowei Xiao
Published: 2026-01-15
Categories: cs.CR, cs.AI, cs.CL
arXiv: https://arxiv.org/abs/2601.10173v1

Large Language Models (LLMs) have enabled the development of powerful agentic systems capable of automating complex workflows across various fields. However, these systems are highly vulnerable to indirect prompt injection attacks, where malicious instructions embedded in external data can hijack agent behavior. In this work, we present ReasAlign, a model-level solution to improve safety alignment against indirect prompt injection attacks. The core idea of ReasAlign is to incorporate structured reasoning steps to analyze user queries, detect conflicting instructions, and preserve the continuity of the user's intended tasks to defend against indirect injection attacks. To further ensure reasoning logic and accuracy, we introduce a test-time scaling mechanism with a preference-optimized judge model that scores reasoning steps and selects the best trajectory. Comprehensive evaluations across various benchmarks show that ReasAlign maintains utility comparable to an undefended model while consistently outperforming Meta SecAlign, the strongest prior guardrail. On the representative open-ended CyberSecEval2 benchmark, which includes multiple prompt-injected tasks, ReasAlign achieves 94.6% utility and only 3.6% ASR, far surpassing the state-of-the-art defensive model of Meta SecAlign (56.4% utility and 74.4% ASR). These results demonstrate that ReasAlign achieves the best trade-off between security and utility, establishing a robust and practical defense against prompt injection attacks in real-world agentic systems. Our code and experimental results could be found at https://github.com/leolee99/ReasAlign.

---

### CtD: Composition through Decomposition in Emergent Communication

Authors: Boaz Carmeli, Ron Meir, Yonatan Belinkov
Published: 2026-01-15
Categories: cs.AI
arXiv: https://arxiv.org/abs/2601.10169v1

Compositionality is a cognitive mechanism that allows humans to systematically combine known concepts in novel ways. This study demonstrates how artificial neural agents acquire and utilize compositional generalization to describe previously unseen images. Our method, termed "Composition through Decomposition", involves two sequential training steps. In the 'Decompose' step, the agents learn to decompose an image into basic concepts using a codebook acquired during interaction in a multi-target coordination game. Subsequently, in the 'Compose' step, the agents employ this codebook to describe novel images by composing basic concepts into complex phrases. Remarkably, we observe cases where generalization in the `Compose' step is achieved zero-shot, without the need for additional training.

---

### RAG-3DSG: Enhancing 3D Scene Graphs with Re-Shot Guided Retrieval-Augmented Generation

Authors: Yue Chang, Rufeng Chen, Zhaofan Zhang, Yi Chen, Yifan Tian, Sihong Xie
Published: 2026-01-15
Categories: cs.CV, cs.AI, cs.RO
arXiv: https://arxiv.org/abs/2601.10168v3

Open-vocabulary 3D Scene Graph (3DSG) can enhance various downstream tasks in robotics by leveraging structured semantic representations, yet current 3DSG construction methods suffer from semantic inconsistencies caused by noisy cross-image aggregation under occlusions and constrained viewpoints. To mitigate the impact of such inconsistency, we propose RAG-3DSG, which introduces re-shot guided uncertainty estimation. By measuring the semantic consistency between original limited viewpoints and re-shot optimal viewpoints, this method quantifies the underlying semantic ambiguity of each graph object. Based on this quantification, we devise an Object-level Retrieval-Augmented Generation (RAG) that leverages low-uncertainty objects as semantic anchors to retrieve more reliable contextual knowledge, enabling a Vision-Language Model to rectify the predictions of uncertain objects and optimize the final 3DSG. Extensive evaluations across three challenging benchmarks and real-world robot trials demonstrate that RAG-3DSG achieves superior recall and precision, effectively mitigating semantic noise to provide highly reliable scene representations for robotics tasks.

---

### Credit C-GPT: A Domain-Specialized Large Language Model for Conversational Understanding in Vietnamese Debt Collection

Authors: Nhung Nguyen Thi Hong, Cuong Nguyen Dang, Tri Le Ngoc
Published: 2026-01-15
Categories: cs.CL
arXiv: https://arxiv.org/abs/2601.10167v1

Debt collection is a critical function within the banking, financial services, and insurance (BFSI) sector, relying heavily on large-scale human-to-human conversational interactions conducted primarily in Vietnamese contact centers. These conversations involve informal spoken language, emotional variability, and complex domain-specific reasoning, which pose significant challenges for traditional natural language processing systems. This paper introduces Credit C-GPT, a domain-specialized large language model with seven billion parameters, fine-tuned for conversational understanding in Vietnamese debt collection scenarios. The proposed model integrates multiple conversational intelligence tasks, including dialogue understanding, sentiment recognition, intent detection, call stage classification, and structured slot-value extraction, within a single reasoning-based framework. We describe the data construction process, annotation strategy, and training methodology, and evaluate the model on proprietary human-annotated datasets. Experimental results show consistent improvements over traditional pipeline-based approaches, indicating that domain-specialized conversational language models provide a scalable and privacy-aware solution for real-time assistance and post-call analytics in enterprise contact centers.

---

### Advancing Adaptive Multi-Stage Video Anomaly Reasoning: A Benchmark Dataset and Method

Authors: Chao Huang, Benfeng Wang, Wei Wang, Jie Wen, Li Shen, Wenqi Ren, Yong Xu, Xiaochun Cao
Published: 2026-01-15
Categories: cs.CV
arXiv: https://arxiv.org/abs/2601.10165v1

Recent progress in reasoning capabilities of Multimodal Large Language Models(MLLMs) has highlighted their potential for performing complex video understanding tasks. However, in the domain of Video Anomaly Detection and Understanding (VAD&U), existing MLLM-based methods are largely limited to anomaly localization or post-hoc description, lacking explicit reasoning processes, risk awareness, and decision-oriented interpretation. To address this gap, we define a new task termed Video Anomaly Reasoning (VAR), which elevates video anomaly analysis from descriptive understanding to structured, multi-stage reasoning. VAR explicitly requires models to perform progressive reasoning over anomalous events before answering anomaly-related questions, encompassing visual perception, causal interpretation, and risk-aware decision making. To support this task, we present a new dataset with 8,641 videos, where each video is annotated with diverse question types corresponding to different reasoning depths, totaling more than 50,000 samples, making it one of the largest datasets for video anomaly. The annotations are based on a structured Perception-Cognition-Action Chain-of-Thought (PerCoAct-CoT), which formalizes domain-specific reasoning priors for video anomaly understanding. This design enables systematic evaluation of multi-stage and adaptive anomaly reasoning. In addition, we propose Anomaly-Aware Group Relative Policy Optimization to further enhance reasoning reliability under weak supervision. Building upon the proposed task and dataset, we develop an end-to-end MLLM-based VAR model termed Vad-R1-Plus, which supports adaptive hierarchical reasoning and risk-aware decision making. Extensive experiments demonstrate that the proposed benchmark and method effectively advance the reasoning capabilities of MLLMs on VAR tasks, outperforming both open-source and proprietary baselines.

---

### AWED-PIPER: Agents, Web Applications & Expert Detectors for Personally Identifiable Information Protection & Fine-grained Named Entity Recognition across 36 languages for 6.6 Billion Speakers

Authors: Prachuryya Kaushik, Ashish Anand
Published: 2026-01-15
Categories: cs.CL, cs.AI, cs.IR
arXiv: https://arxiv.org/abs/2601.10161v3

Named Entity Recognition (NER) and Personally Identifiable Information (PII) anonymization are critical tasks in Natural Language Processing (NLP) for information extraction and privacy preservation. We introduce AWED-PIPER, an open-source framework comprising agentic tools, interactive web applications, and 54 state-of-the-art expert detector models that provide unified Fine-grained Named Entity Recognition (FgNER) and reversible synthetic PII pseudonymization across 36 languages spoken by over 6.6 billion people. The system couples fine-grained multilingual sequence labeling with script-aware regex detectors to identify contextual entities (Person, Location, Organization, Medical) as well as structured technical PII (Emails, native-script Phone Numbers, IP Addresses, Credit Cards). AWED-PIPER offers a dual capability: full FgNER entity extraction and privacy-preserving reversible anonymization with persistent placeholders and de-anonymization dictionary mappings. The suite spans global languages to extremely low-resource vulnerable languages like Bodo, Manipuri, Bishnupriya, and Mizo. The resources can be accessed here: PII Protector Agentic Tool: (https://github.com/PrachuryyaKaushik/AWED-PIPER), FgNER Agentic Tool: (https://github.com/PrachuryyaKaushik/AWED-FiNER), PII Web Application: (https://hf.co/spaces/prachuryyaIITG/AWED_PII_Protector), FgNER Web Application: (https://hf.co/spaces/prachuryyaIITG/AWED-FiNER), and Edge-deployable Expert Detector Models: (https://hf.co/collections/prachuryyaIITG/awed-piper).

---

### Alignment Pretraining: AI Discourse Causes Self-Fulfilling (Mis)alignment

Authors: Cameron Tice, Puria Radmard, Samuel Ratnam, Andy Kim, David Africa, Kyle O'Brien
Published: 2026-01-15
Categories: cs.CL, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2601.10160v2

Pretraining corpora contain extensive discourse about AI systems, yet the causal influence of this discourse on downstream alignment remains poorly understood. If prevailing descriptions of AI behaviour are predominantly negative, LLMs may internalise corresponding behavioural priors, giving rise to self-fulfilling misalignment. This paper provides the first controlled study of this hypothesis by pretraining 6.9B-parameter LLMs with varying amounts of (mis)alignment discourse. We find that discussion of AI contributes to misalignment. Upsampling synthetic training documents about AI misalignment leads to a notable increase in misaligned behaviour. Conversely, upsampling documents about aligned behaviour reduces misalignment scores from 45% to 9%. We consider this evidence of self-fulfilling alignment. These effects are dampened, but persist through post-training. Our findings establish the study of how pretraining data shapes alignment priors, or alignment pretraining, as a complement to post-training. We recommend practitioners consider pretraining for alignment alongside capabilities. We share our models, data, and evaluations at AlignmentPretraining.ai.

---

### What Gets Activated: Uncovering Domain and Driver Experts in MoE Language Models

Authors: Guimin Hu, Meng Li, Qiwei Peng, Lijie Hu, Boyan Xu, Ruichu Cai
Published: 2026-01-15
Categories: cs.CL
arXiv: https://arxiv.org/abs/2601.10159v2

Most interpretability work focuses on layer- or neuron-level mechanisms in Transformers, leaving expert-level behavior in MoE LLMs underexplored. Motivated by functional specialization in the human brain, we analyze expert activation by distinguishing domain and driver experts. In this work, we study expert activation in MoE models across three public domains and address two key questions: (1) which experts are activated, and whether certain expert types exhibit consistent activation patterns; and (2) how tokens are associated with and trigger the activation of specific experts. To answer these questions, we introduce entropy-based and causal-effect metrics to assess whether an expert is strongly favored for a particular domain, and how strongly expert activation contributes causally to the model's output, thus identify domain and driver experts, respectively. Furthermore, we explore how individual tokens are associated with the activation of specific experts. Our analysis reveals that (1) Among the activated experts, some show clear domain preferences, while others exert strong causal influence on model performance, underscoring their decisive roles. (2) tokens occurring earlier in a sentence are more likely to trigger the driver experts, and (3) adjusting the weights of domain and driver experts leads to significant performance gains across all three models and domains. These findings shed light on the internal mechanisms of MoE models and enhance their interpretability.

---

### MMPG: MoE-based Adaptive Multi-Perspective Graph Fusion for Protein Representation Learning

Authors: Yusong Wang, Jialun Shen, Zhihao Wu, Yicheng Xu, Shiyin Tan, Mingkun Xu, Changshuo Wang, Zixing Song, Prayag Tiwari
Published: 2026-01-15
Categories: cs.AI
arXiv: https://arxiv.org/abs/2601.10157v1

Graph Neural Networks (GNNs) have been widely adopted for Protein Representation Learning (PRL), as residue interaction networks can be naturally represented as graphs. Current GNN-based PRL methods typically rely on single-perspective graph construction strategies, which capture partial properties of residue interactions, resulting in incomplete protein representations. To address this limitation, we propose MMPG, a framework that constructs protein graphs from multiple perspectives and adaptively fuses them via Mixture of Experts (MoE) for PRL. MMPG constructs graphs from physical, chemical, and geometric perspectives to characterize different properties of residue interactions. To capture both perspective-specific features and their synergies, we develop an MoE module, which dynamically routes perspectives to specialized experts, where experts learn intrinsic features and cross-perspective interactions. We quantitatively verify that MoE automatically specializes experts in modeling distinct levels of interaction from individual representations, to pairwise inter-perspective synergies, and ultimately to a global consensus across all perspectives. Through integrating this multi-level information, MMPG produces superior protein representations and achieves advanced performance on four different downstream protein tasks.

---

### ToolSafe: Enhancing Tool Invocation Safety of LLM-based agents via Proactive Step-level Guardrail and Feedback

Authors: Yutao Mou, Zhangchi Xue, Lijun Li, Peiyang Liu, Shikun Zhang, Wei Ye, Jing Shao
Published: 2026-01-15
Categories: cs.CL
arXiv: https://arxiv.org/abs/2601.10156v1

While LLM-based agents can interact with environments via invoking external tools, their expanded capabilities also amplify security risks. Monitoring step-level tool invocation behaviors in real time and proactively intervening before unsafe execution is critical for agent deployment, yet remains under-explored. In this work, we first construct TS-Bench, a novel benchmark for step-level tool invocation safety detection in LLM agents. We then develop a guardrail model, TS-Guard, using multi-task reinforcement learning. The model proactively detects unsafe tool invocation actions before execution by reasoning over the interaction history. It assesses request harmfulness and action-attack correlations, producing interpretable and generalizable safety judgments and feedback. Furthermore, we introduce TS-Flow, a guardrail-feedback-driven reasoning framework for LLM agents, which reduces harmful tool invocations of ReAct-style agents by 65 percent on average and improves benign task completion by approximately 10 percent under prompt injection attacks.

---

### LOOKAT: Lookup-Optimized Key-Attention for Memory-Efficient Transformers

Authors: Aryan Karmore
Published: 2026-01-15
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2601.10155v1

Compressing the KV cache is a required step to deploy large language models on edge devices. Current quantization methods compress storage but fail to reduce bandwidth as attention calculation requires dequantizing keys from INT4/INT8 to FP16 before use. We observe that attention scoring is mathematically equivalent to the inner product similarity search and we can apply some compression techniques from vector databases to compress KV-cache better. We propose LOOKAT, which applies product quantization and asymmetric distance computation, to transformer architecture by decomposing key vectors into subspaces, learning codebooks and computing attention tables via lookup tables. This transforms attention from memory-bound to compute-bound. LOOKAT achieves 64 $\times$ compression at 95.7\% output fidelity and 32 $\times$ compression at 95.0\% fidelity when tested on GPT-2. LOOKAT requires no architecture changes or training while maintaining rank correlation $ρ> 0.95$. Theoretical analysis confirms that rank correlation degrades as $O(d_k/mK)$, with guarantees validated across sequence lengths up to 1024 tokens.

---

### MHub.ai: A Simple, Standardized, and Reproducible Platform for AI Models in Medical Imaging

Authors: Leonard Nürnberg, Dennis Bontempi, Suraj Pai, Curtis Lisle, Steve Pieper, Ron Kikinis, Sil van de Leemput, Rahul Soni, Gowtham Murugesan, Cosmin Ciausu, Miriam Groeneveld, Felix J. Dorfner et al.
Published: 2026-01-15
Categories: cs.AI, cs.CV, cs.ET, cs.LG, cs.SE
arXiv: https://arxiv.org/abs/2601.10154v1

Artificial intelligence (AI) has the potential to transform medical imaging by automating image analysis and accelerating clinical research. However, research and clinical use are limited by the wide variety of AI implementations and architectures, inconsistent documentation, and reproducibility issues. Here, we introduce MHub$.$ai, an open-source, container-based platform that standardizes access to AI models with minimal configuration, promoting accessibility and reproducibility in medical imaging. MHub$.$ai packages models from peer-reviewed publications into standardized containers that support direct processing of DICOM and other formats, provide a unified application interface, and embed structured metadata. Each model is accompanied by publicly available reference data that can be used to confirm model operation. MHub$.$ai includes an initial set of state-of-the-art segmentation, prediction, and feature extraction models for different modalities. The modular framework enables adaptation of any model and supports community contributions. We demonstrate the utility of the platform in a clinical use case through comparative evaluation of lung segmentation models. To further strengthen transparency and reproducibility, we publicly release the generated segmentations and evaluation metrics and provide interactive dashboards that allow readers to inspect individual cases and reproduce or extend our analysis. By simplifying model use, MHub$.$ai enables side-by-side benchmarking with identical execution commands and standardized outputs, and lowers the barrier to clinical translation.

---

### Simple Network Graph Comparative Learning

Authors: Qiang Yu, Xinran Cheng, Shiqiang Xu, Chuanyi Liu
Published: 2026-01-15
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2601.10150v1

The effectiveness of contrastive learning methods has been widely recognized in the field of graph learning, especially in contexts where graph data often lack labels or are difficult to label. However, the application of these methods to node classification tasks still faces a number of challenges. First, existing data enhancement techniques may lead to significant differences from the original view when generating new views, which may weaken the relevance of the view and affect the efficiency of model training. Second, the vast majority of existing graph comparison learning algorithms rely on the use of a large number of negative samples. To address the above challenges, this study proposes a novel node classification contrast learning method called Simple Network Graph Comparative Learning (SNGCL). Specifically, SNGCL employs a superimposed multilayer Laplace smoothing filter as a step in processing the data to obtain global and local feature smoothing matrices, respectively, which are thus passed into the target and online networks of the siamese network, and finally employs an improved triple recombination loss function to bring the intra-class distance closer and the inter-class distance farther. We have compared SNGCL with state-of-the-art models in node classification tasks, and the experimental results show that SNGCL is strongly competitive in most tasks.

---

### DecisionLLM: Large Language Models for Long Sequence Decision Exploration

Authors: Xiaowei Lv, Zhilin Zhang, Yijun Li, Yusen Huo, Siyuan Ju, Xuyan Li, Chunxiang Hong, Tianyu Wang, Yongcai Wang, Peng Sun, Chuan Yu, Jian Xu et al.
Published: 2026-01-15
Categories: cs.AI
arXiv: https://arxiv.org/abs/2601.10148v1

Long-sequence decision-making, which is usually addressed through reinforcement learning (RL), is a critical component for optimizing strategic operations in dynamic environments, such as real-time bidding in computational advertising. The Decision Transformer (DT) introduced a powerful paradigm by framing RL as an autoregressive sequence modeling problem. Concurrently, Large Language Models (LLMs) have demonstrated remarkable success in complex reasoning and planning tasks. This inspires us whether LLMs, which share the same Transformer foundation, but operate at a much larger scale, can unlock new levels of performance in long-horizon sequential decision-making problem. This work investigates the application of LLMs to offline decision making tasks. A fundamental challenge in this domain is the LLMs' inherent inability to interpret continuous values, as they lack a native understanding of numerical magnitude and order when values are represented as text strings. To address this, we propose treating trajectories as a distinct modality. By learning to align trajectory data with natural language task descriptions, our model can autoregressively predict future decisions within a cohesive framework we term DecisionLLM. We establish a set of scaling laws governing this paradigm, demonstrating that performance hinges on three factors: model scale, data volume, and data quality. In offline experimental benchmarks and bidding scenarios, DecisionLLM achieves strong performance. Specifically, DecisionLLM-3B outperforms the traditional Decision Transformer (DT) by 69.4 on Maze2D umaze-v1 and by 0.085 on AuctionNet. It extends the AIGB paradigm and points to promising directions for future exploration in online bidding.

---

### History Is Not Enough: An Adaptive Dataflow System for Financial Time-Series Synthesis

Authors: Haochong Xia, Yao Long Teng, Regan Tan, Molei Qin, Xinrun Wang, Bo An
Published: 2026-01-15
Categories: cs.AI, q-fin.TR
arXiv: https://arxiv.org/abs/2601.10143v1

In quantitative finance, the gap between training and real-world performance-driven by concept drift and distributional non-stationarity-remains a critical obstacle for building reliable data-driven systems. Models trained on static historical data often overfit, resulting in poor generalization in dynamic markets. The mantra "History Is Not Enough" underscores the need for adaptive data generation that learns to evolve with the market rather than relying solely on past observations. We present a drift-aware dataflow system that integrates machine learning-based adaptive control into the data curation process. The system couples a parameterized data manipulation module comprising single-stock transformations, multi-stock mix-ups, and curation operations, with an adaptive planner-scheduler that employs gradient-based bi-level optimization to control the system. This design unifies data augmentation, curriculum learning, and data workflow management under a single differentiable framework, enabling provenance-aware replay and continuous data quality monitoring. Extensive experiments on forecasting and reinforcement learning trading tasks demonstrate that our framework enhances model robustness and improves risk-adjusted returns. The system provides a generalizable approach to adaptive data management and learning-guided workflow automation for financial data.

---

### Understanding and Preserving Safety in Fine-Tuned LLMs

Authors: Jiawen Zhang, Yangfan Hu, Kejia Chen, Lipeng He, Jiachen Ma, Jian Lou, Dan Li, Jian Liu, Xiaohu Yang, Ruoxi Jia
Published: 2026-01-15
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2601.10141v1

Fine-tuning is an essential and pervasive functionality for applying large language models (LLMs) to downstream tasks. However, it has the potential to substantially degrade safety alignment, e.g., by greatly increasing susceptibility to jailbreak attacks, even when the fine-tuning data is entirely harmless. Despite garnering growing attention in defense efforts during the fine-tuning stage, existing methods struggle with a persistent safety-utility dilemma: emphasizing safety compromises task performance, whereas prioritizing utility typically requires deep fine-tuning that inevitably leads to steep safety declination. In this work, we address this dilemma by shedding new light on the geometric interaction between safety- and utility-oriented gradients in safety-aligned LLMs. Through systematic empirical analysis, we uncover three key insights: (I) safety gradients lie in a low-rank subspace, while utility gradients span a broader high-dimensional space; (II) these subspaces are often negatively correlated, causing directional conflicts during fine-tuning; and (III) the dominant safety direction can be efficiently estimated from a single sample. Building upon these novel insights, we propose safety-preserving fine-tuning (SPF), a lightweight approach that explicitly removes gradient components conflicting with the low-rank safety subspace. Theoretically, we show that SPF guarantees utility convergence while bounding safety drift. Empirically, SPF consistently maintains downstream task performance and recovers nearly all pre-trained safety alignment, even under adversarial fine-tuning scenarios. Furthermore, SPF exhibits robust resistance to both deep fine-tuning and dynamic jailbreak attacks. Together, our findings provide new mechanistic understanding and practical guidance toward always-aligned LLM fine-tuning.

---

### Step-by-Step Causality: Transparent Causal Discovery with Multi-Agent Tree-Query and Adversarial Confidence Estimation

Authors: Ziyi Ding, Chenfei Ye-Hao, Zheyuan Wang, Xiao-Ping Zhang
Published: 2026-01-15
Categories: cs.LG, cs.AI, stat.ML
arXiv: https://arxiv.org/abs/2601.10137v1

Causal discovery aims to recover ``what causes what'', but classical constraint-based methods (e.g., PC, FCI) suffer from error propagation, and recent LLM-based causal oracles often behave as opaque, confidence-free black boxes. This paper introduces Tree-Query, a tree-structured, multi-expert LLM framework that reduces pairwise causal discovery to a short sequence of queries about backdoor paths, (in)dependence, latent confounding, and causal direction, yielding interpretable judgments with robustness-aware confidence scores. Theoretical guarantees are provided for asymptotic identifiability of four pairwise relations. On data-free benchmarks derived from Mooij et al. and UCI causal graphs, Tree-Query improves structural metrics over direct LLM baselines, and a diet--weight case study illustrates confounder screening and stable, high-confidence causal conclusions. Tree-Query thus offers a principled way to obtain data-free causal priors from LLMs that can complement downstream data-driven causal discovery. Code is available at https://anonymous.4open.science/r/Repo-9B3E-4F96.

---

### Is More Context Always Better? Examining LLM Reasoning Capability for Time Interval Prediction

Authors: Yanan Cao, Farnaz Fallahi, Murali Mohana Krishna Dandu, Lalitesh Morishetti, Kai Zhao, Luyi Ma, Sinduja Subramaniam, Jianpeng Xu, Evren Korpeoglu, Kaushiki Nag, Sushant Kumar, Kannan Achan
Published: 2026-01-15
Categories: cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2601.10132v2

Large Language Models (LLMs) have demonstrated impressive capabilities in reasoning and prediction across different domains. Yet, their ability to infer temporal regularities from structured behavioral data remains underexplored. This paper presents a systematic study investigating whether LLMs can predict time intervals between recurring user actions, such as repeated purchases, and how different levels of contextual information shape their predictive behavior. Using a simple but representative repurchase scenario, we benchmark state-of-the-art LLMs in zero-shot settings against both statistical and machine-learning models. Two key findings emerge. First, while LLMs surpass lightweight statistical baselines, they consistently underperform dedicated machine-learning models, showing their limited ability to capture quantitative temporal structure. Second, although moderate context can improve LLM accuracy, adding further user-level detail degrades performance. These results challenge the assumption that "more context leads to better reasoning". Our study highlights fundamental limitations of today's LLMs in structured temporal inference and offers guidance for designing future context-aware hybrid models that integrate statistical precision with linguistic flexibility.

---

### M^4olGen: Multi-Agent, Multi-Stage Molecular Generation under Precise Multi-Property Constraints

Authors: Yizhan Li, Florence Cloutier, Sifan Wu, Ali Parviz, Boris Knyazev, Yan Zhang, Glen Berseth, Bang Liu
Published: 2026-01-15
Categories: cs.AI, cs.MA
arXiv: https://arxiv.org/abs/2601.10131v2

Generating molecules that satisfy precise numeric constraints over multiple physicochemical properties is critical and challenging. Although large language models (LLMs) are expressive, they struggle with precise multi-objective control and numeric reasoning without external structure and feedback. We introduce \textbf{M olGen}, a fragment-level, retrieval-augmented, two-stage framework for molecule generation under multi-property constraints. Stage I : Prototype generation: a multi-agent reasoner performs retrieval-anchored, fragment-level edits to produce a candidate near the feasible region. Stage II : RL-based fine-grained optimization: a fragment-level optimizer trained with Group Relative Policy Optimization (GRPO) applies one- or multi-hop refinements to explicitly minimize the property errors toward our target while regulating edit complexity and deviation from the prototype. A large, automatically curated dataset with reasoning chains of fragment edits and measured property deltas underpins both stages, enabling deterministic, reproducible supervision and controllable multi-hop reasoning. Unlike prior work, our framework better reasons about molecules by leveraging fragments and supports controllable refinement toward numeric targets. Experiments on generation under two sets of property constraints (QED, LogP, Molecular Weight and HOMO, LUMO) show consistent gains in validity and precise satisfaction of multi-property targets, outperforming strong LLMs and graph-based algorithms.

---

### Redundancy-Driven Top-$k$ Functional Dependency Discovery

Authors: Xiaolong Wan, Xixian Han
Published: 2026-01-15
Categories: cs.DB, cs.AI
arXiv: https://arxiv.org/abs/2601.10130v1

Functional dependencies (FDs) are basic constraints in relational databases and are used for many data management tasks. Most FD discovery algorithms find all valid dependencies, but this causes two problems. First, the computational cost is prohibitive: computational complexity grows quadratically with the number of tuples and exponentially with the number of attributes, making discovery slow on large-scale and high-dimensional data. Second, the result set can be huge, making it hard to identify useful dependencies. We propose SDP (Selective-Discovery-and-Prune), which discovers the top-$k$ FDs ranked by redundancy count. Redundancy count measures how much duplicated information an FD explains and connects directly to storage overhead and update anomalies. SDP uses an upper bound on redundancy to prune the search space. It is proved that this upper bound is monotone: adding attributes refines partitions and thus decreases the bound. Once the bound falls below the top-$k$ threshold, the entire branch can be skipped. We improve SDP with three optimizations: ordering attributes by partition cardinality, using pairwise statistics in a Partition Cardinality Matrix to tighten bounds, and a global scheduler to explore promising branches first. Experiments on over 40 datasets show that SDP is much faster and uses less memory than exhaustive methods.

---

### LaViT: Aligning Latent Visual Thoughts for Multi-modal Reasoning

Authors: Linquan Wu, Tianxiang Jiang, Yifei Dong, Haoyu Yang, Fengji Zhang, Shichaang Meng, Ai Xuan, Linqi Song, Jacky Keung
Published: 2026-01-15
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2601.10129v2

Current multimodal latent reasoning often relies on external supervision (e.g., auxiliary images), ignoring intrinsic visual attention dynamics. In this work, we identify a critical Perception Gap in distillation: student models frequently mimic a teacher's textual output while attending to fundamentally divergent visual regions, effectively relying on language priors rather than grounded perception. To bridge this, we propose LaViT, a framework that aligns latent visual thoughts rather than static embeddings. LaViT compels the student to autoregressively reconstruct the teacher's visual semantics and attention trajectories prior to text generation, employing a curriculum sensory gating mechanism to prevent shortcut learning. Extensive experiments show that LaViT significantly enhances visual grounding, achieving up to +16.9% gains on complex reasoning tasks and enabling a compact 3B model to outperform larger open-source variants and proprietary models like GPT-4o.

---

### VQ-Seg: Vector-Quantized Token Perturbation for Semi-Supervised Medical Image Segmentation

Authors: Sicheng Yang, Zhaohu Xing, Lei Zhu
Published: 2026-01-15
Categories: cs.CV
arXiv: https://arxiv.org/abs/2601.10124v1

Consistency learning with feature perturbation is a widely used strategy in semi-supervised medical image segmentation. However, many existing perturbation methods rely on dropout, and thus require a careful manual tuning of the dropout rate, which is a sensitive hyperparameter and often difficult to optimize and may lead to suboptimal regularization. To overcome this limitation, we propose VQ-Seg, the first approach to employ vector quantization (VQ) to discretize the feature space and introduce a novel and controllable Quantized Perturbation Module (QPM) that replaces dropout. Our QPM perturbs discrete representations by shuffling the spatial locations of codebook indices, enabling effective and controllable regularization. To mitigate potential information loss caused by quantization, we design a dual-branch architecture where the post-quantization feature space is shared by both image reconstruction and segmentation tasks. Moreover, we introduce a Post-VQ Feature Adapter (PFA) to incorporate guidance from a foundation model (FM), supplementing the high-level semantic information lost during quantization. Furthermore, we collect a large-scale Lung Cancer (LC) dataset comprising 828 CT scans annotated for central-type lung carcinoma. Extensive experiments on the LC dataset and other public benchmarks demonstrate the effectiveness of our method, which outperforms state-of-the-art approaches. Code available at: https://github.com/script-Yang/VQ-Seg.

---

### Role-Playing Agents Driven by Large Language Models: Current Status, Challenges, and Future Trends

Authors: Ye Wang, Jiaxing Chen, Hongjiang Xiao
Published: 2026-01-15
Categories: cs.CL, cs.AI, cs.HC
arXiv: https://arxiv.org/abs/2601.10122v1

In recent years, with the rapid advancement of large language models (LLMs), role-playing language agents (RPLAs) have emerged as a prominent research focus at the intersection of natural language processing (NLP) and human-computer interaction. This paper systematically reviews the current development and key technologies of RPLAs, delineating the technological evolution from early rule-based template paradigms, through the language style imitation stage, to the cognitive simulation stage centered on personality modeling and memory mechanisms. It summarizes the critical technical pathways supporting high-quality role-playing, including psychological scale-driven character modeling, memory-augmented prompting mechanisms, and motivation-situation-based behavioral decision control. At the data level, the paper further analyzes the methods and challenges of constructing role-specific corpora, focusing on data sources, copyright constraints, and structured annotation processes. In terms of evaluation, it collates multi-dimensional assessment frameworks and benchmark datasets covering role knowledge, personality fidelity, value alignment, and interactive hallucination, while commenting on the advantages and disadvantages of methods such as human evaluation, reward models, and LLM-based scoring. Finally, the paper outlines future development directions of role-playing agents, including personality evolution modeling, multi-agent collaborative narrative, multimodal immersive interaction, and integration with cognitive neuroscience, aiming to provide a systematic perspective and methodological insights for subsequent research.

---

### TopoDIM: One-shot Topology Generation of Diverse Interaction Modes for Multi-Agent Systems

Authors: Rui Sun, Jie Ding, Chenghua Gong, Tianjun Gu, Yihang Jiang, Juyuan Zhang, Liming Pan, Linyuan Lü
Published: 2026-01-15
Categories: cs.MA, cs.AI, cs.CL
arXiv: https://arxiv.org/abs/2601.10120v2

Optimizing communication topology in LLM-based multi-agent system is critical for enabling collective intelligence. Existing methods mainly rely on spatio-temporal interaction paradigms, where the sequential execution of multi-round dialogues incurs high latency and computation. Motivated by the recent insights that evaluation and debate mechanisms can improve problem-solving in multi-agent systems, we propose TopoDIM, a framework for one-shot Topology generation with Diverse Interaction Modes. Designed for decentralized execution to enhance adaptability and privacy, TopoDIM enables agents to autonomously construct heterogeneous communication without iterative coordination, achieving token efficiency and improved task performance. Experiments demonstrate that TopoDIM reduces total token consumption by 46.41% while improving average performance by 1.50% over state-of-the-art methods. Moreover, the framework exhibits strong adaptability in organizing communication among heterogeneous agents. Code is available at: https://github.com/Sundiasy/TopoDIM.

---

### Beyond Single Prompts: Synergistic Fusion and Arrangement for VICL

Authors: Wenwen Liao, Jianbo Yu, Yuansong Wang, Shifu Yan, Xiaofeng Yang
Published: 2026-01-15
Categories: cs.CV
arXiv: https://arxiv.org/abs/2601.10117v1

Vision In-Context Learning (VICL) enables inpainting models to quickly adapt to new visual tasks from only a few prompts. However, existing methods suffer from two key issues: (1) selecting only the most similar prompt discards complementary cues from other high-quality prompts; and (2) failing to exploit the structured information implied by different prompt arrangements. We propose an end-to-end VICL framework to overcome these limitations. Firstly, an adaptive Fusion Module aggregates critical patterns and annotations from multiple prompts to form more precise contextual prompts. Secondly, we introduce arrangement-specific lightweight MLPs to decouple layout priors from the core model, while minimally affecting the overall model. In addition, an bidirectional fine-tuning mechanism swaps the roles of query and prompt, encouraging the model to reconstruct the original prompt from fused context and thus enhancing collaboration between the fusion module and the inpainting model. Experiments on foreground segmentation, single-object detection, and image colorization demonstrate superior results and strong cross-task generalization of our method.

---

### Following the Teacher's Footsteps: Scheduled Checkpoint Distillation for Domain-Specific LLMs

Authors: Cheng Feng, Chaoliang Zhong, Jun Sun, Yusuke Oishi
Published: 2026-01-15
Categories: cs.AI
arXiv: https://arxiv.org/abs/2601.10114v1

Large language models (LLMs) are challenging to deploy for domain-specific tasks due to their massive scale. While distilling a fine-tuned LLM into a smaller student model is a promising alternative, the capacity gap between teacher and student often leads to suboptimal performance. This raises a key question: when and how can a student model match or even surpass its teacher on domain-specific tasks? In this work, we propose a novel theoretical insight: a student can outperform its teacher if its advantage on a Student-Favored Subdomain (SFS) outweighs its deficit on the Teacher-Favored Subdomain (TFS). Guided by this insight, we propose Scheduled Checkpoint Distillation (SCD), which reduces the TFS deficit by emulating the teacher's convergence process during supervised fine-tuning (SFT) on the domain task, and a sample-wise Adaptive Weighting (AW) mechanism to preserve student strengths on SFS. Experiments across diverse domain tasks--including QA, NER, and text classification in multiple languages--show that our method consistently outperforms existing distillation approaches, allowing the student model to match or even exceed the performance of its fine-tuned teacher.

---

### Repository Intelligence Graph: Deterministic Architectural Map for LLM Code Assistants

Authors: Tsvi Cherny-Shahar, Amiram Yehudai
Published: 2026-01-15
Categories: cs.SE, cs.AI
arXiv: https://arxiv.org/abs/2601.10112v1

Repository aware coding agents often struggle to recover build and test structure, especially in multilingual projects where cross language dependencies are encoded across heterogeneous build systems and tooling. We introduce the Repository Intelligence Graph (RIG), a deterministic, evidence backed architectural map that represents buildable components, aggregators, runners, tests, external packages, and package managers, connected by explicit dependency and coverage edges that trace back to concrete build and test definitions. We also present SPADE, a deterministic extractor that constructs RIG from build and test artifacts (currently with an automatic CMake plugin based on the CMake File API and CTest metadata), and exposes RIG as an LLM friendly JSON view that agents can treat as the authoritative description of repository structure. We evaluate three commercial agents (Claude Code, Cursor, Codex) on eight repositories spanning low to high build oriented complexity, including the real world MetaFFI project. Each agent answers thirty structured questions per repository with and without RIG in context, and we measure accuracy, wall clock completion time, and efficiency (seconds per correct answer). Across repositories and agents, providing RIG improves mean accuracy by 12.2\% and reduces completion time by 53.9\%, yielding a mean 57.8\% reduction in seconds per correct answer. Gains are larger in multilingual repositories, which improve by 17.7\% in accuracy and 69.5\% in efficiency on average, compared to 6.6\% and 46.1\% in single language repositories. Qualitative analysis suggests that RIG shifts failures from structural misunderstandings toward reasoning mistakes over a correct structure, while rare regressions highlight that graph based reasoning quality remains a key factor.

---

### Multi-Constrained Evolutionary Molecular Design Framework: An Interpretable Drug Design Method Combining Rule-Based Evolution and Molecular Crossover

Authors: Shanxian Lin, Wei Xia, Yuichi Nagata, Haichuan Yang
Published: 2026-01-15
Categories: cs.NE
arXiv: https://arxiv.org/abs/2601.10110v1

This study proposes MCEMOL (Multi-Constrained Evolutionary Molecular Design Framework), a molecular optimization approach integrating rule-based evolution with molecular crossover. MCEMOL employs dual-layer evolution: optimizing transformation rules at rule level while applying crossover and mutation to molecular structures. Unlike deep learning methods requiring large datasets and extensive training, our algorithm evolves efficiently from minimal starting molecules with low computational overhead. The framework incorporates message-passing neural networks and comprehensive chemical constraints, ensuring efficient and interpretable molecular design. Experimental results demonstrate that MCEMOL provides transparent design pathways through its evolutionary mechanism while generating valid, diverse, target-compliant molecules. The framework achieves 100% molecular validity with high structural diversity and excellent drug-likeness compliance, showing strong performance in symmetry constraints, pharmacophore optimization, and stereochemical integrity. Unlike black-box methods, MCEMOL delivers dual value: interpretable transformation rules researchers can understand and trust, alongside high-quality molecular libraries for practical applications. This establishes a paradigm where interpretable AI-driven drug design and effective molecular generation are achieved simultaneously, bridging the gap between computational innovation and practical drug discovery needs.

---

### Skill-Aware Data Selection and Fine-Tuning for Data-Efficient Reasoning Distillation

Authors: Lechen Zhang, Yunxiang Zhang, Wei Hu, Lu Wang
Published: 2026-01-15
Categories: cs.CL
arXiv: https://arxiv.org/abs/2601.10109v1

Large reasoning models such as DeepSeek-R1 and their distilled variants achieve strong performance on complex reasoning tasks. Yet, distilling these models often demands large-scale data for supervised fine-tuning (SFT), motivating the pursuit of data-efficient training methods. To address this, we propose a skill-centric distillation framework that efficiently transfers reasoning ability to weaker models with two components: (1) Skill-based data selection, which prioritizes examples targeting the student model's weaker skills, and (2) Skill-aware fine-tuning, which encourages explicit skill decomposition during problem solving. With only 1,000 training examples selected from a 100K teacher-generated corpus, our method surpasses random SFT baselines by +1.6% on Qwen3-4B and +1.4% on Qwen3-8B across five mathematical reasoning benchmarks. Further analysis confirms that these gains concentrate on skills emphasized during training, highlighting the effectiveness of skill-centric training for efficient reasoning distillation.

---

### SIN-Bench: Tracing Native Evidence Chains in Long-Context Multimodal Scientific Interleaved Literature

Authors: Yiming Ren, Junjie Wang, Yuxin Meng, Yihang Shi, Zhiqiang Lin, Ruihang Chu, Yiran Xu, Ziming Li, Yunfei Zhao, Zihan Wang, Yu Qiao, Ruiming Tang et al.
Published: 2026-01-15
Categories: cs.CL, cs.AI, cs.MM
arXiv: https://arxiv.org/abs/2601.10108v1

Evaluating whether multimodal large language models truly understand long-form scientific papers remains challenging: answer-only metrics and synthetic "Needle-In-A-Haystack" tests often reward answer matching without requiring a causal, evidence-linked reasoning trace in the document. We propose the "Fish-in-the-Ocean" (FITO) paradigm, which requires models to construct explicit cross-modal evidence chains within native scientific documents. To operationalize FITO, we build SIN-Data, a scientific interleaved corpus that preserves the native interleaving of text and figures. On top of it, we construct SIN-Bench with four progressive tasks covering evidence discovery (SIN-Find), hypothesis verification (SIN-Verify), grounded QA (SIN-QA), and evidence-anchored synthesis (SIN-Summary). We further introduce "No Evidence, No Score", scoring predictions when grounded to verifiable anchors and diagnosing evidence quality via matching, relevance, and logic. Experiments on eight MLLMs show that grounding is the primary bottleneck: Gemini-3-pro achieves the best average overall score (0.573), while GPT-5 attains the highest SIN-QA answer accuracy (0.767) but underperforms on evidence-aligned overall scores, exposing a gap between correctness and traceable support.

---

### Enhancing Visual In-Context Learning by Multi-Faceted Fusion

Authors: Wenwen Liao, Jianbo Yu, Yuansong Wang, Qingchao Jiang, Xiaofeng Yang
Published: 2026-01-15
Categories: cs.CV
arXiv: https://arxiv.org/abs/2601.10107v1

Visual In-Context Learning (VICL) has emerged as a powerful paradigm, enabling models to perform novel visual tasks by learning from in-context examples. The dominant "retrieve-then-prompt" approach typically relies on selecting the single best visual prompt, a practice that often discards valuable contextual information from other suitable candidates. While recent work has explored fusing the top-K prompts into a single, enhanced representation, this still simply collapses multiple rich signals into one, limiting the model's reasoning capability. We argue that a more multi-faceted, collaborative fusion is required to unlock the full potential of these diverse contexts. To address this limitation, we introduce a novel framework that moves beyond single-prompt fusion towards an multi-combination collaborative fusion. Instead of collapsing multiple prompts into one, our method generates three contextual representation branches, each formed by integrating information from different combinations of top-quality prompts. These complementary guidance signals are then fed into proposed MULTI-VQGAN architecture, which is designed to jointly interpret and utilize collaborative information from multiple sources. Extensive experiments on diverse tasks, including foreground segmentation, single-object detection, and image colorization, highlight its strong cross-task generalization, effective contextual fusion, and ability to produce more robust and accurate predictions than existing methods.

---

### MathDoc: Benchmarking Structured Extraction and Active Refusal on Noisy Mathematics Exam Papers

Authors: Chenyue Zhou, Jiayi Tuo, Shitong Qin, Wei Dai, Mingxuan Wang, Ziwei Zhao, Duoyang Li, Shiyang Su, Yanxi Lu, Yanbiao Ma
Published: 2026-01-15
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2601.10104v1

The automated extraction of structured questions from paper-based mathematics exams is fundamental to intelligent education, yet remains challenging in real-world settings due to severe visual noise. Existing benchmarks mainly focus on clean documents or generic layout analysis, overlooking both the structural integrity of mathematical problems and the ability of models to actively reject incomplete inputs. We introduce MathDoc, the first benchmark for document-level information extraction from authentic high school mathematics exam papers. MathDoc contains \textbf{3,609} carefully curated questions with real-world artifacts and explicitly includes unrecognizable samples to evaluate active refusal behavior. We propose a multi-dimensional evaluation framework covering stem accuracy, visual similarity, and refusal capability. Experiments on SOTA MLLMs, including Qwen3-VL and Gemini-2.5-Pro, show that although end-to-end models achieve strong extraction performance, they consistently fail to refuse illegible inputs, instead producing confident but invalid outputs. These results highlight a critical gap in current MLLMs and establish MathDoc as a benchmark for assessing model reliability under degraded document conditions. Our project repository is available at \href{https://github.com/winnk123/papers/tree/master}{GitHub repository}

---

### FlowAct-R1: Towards Interactive Humanoid Video Generation

Authors: Lizhen Wang, Yongming Zhu, Zhipeng Ge, Youwei Zheng, Longhao Zhang, Tianshu Hu, Shiyang Qin, Mingshuang Luo, Jiaxu Zhang, Xin Chen, Yulong Wang, Zerong Zheng et al.
Published: 2026-01-15
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2601.10103v1

Interactive humanoid video generation aims to synthesize lifelike visual agents that can engage with humans through continuous and responsive video. Despite recent advances in video synthesis, existing methods often grapple with the trade-off between high-fidelity synthesis and real-time interaction requirements. In this paper, we propose FlowAct-R1, a framework specifically designed for real-time interactive humanoid video generation. Built upon a MMDiT architecture, FlowAct-R1 enables the streaming synthesis of video with arbitrary durations while maintaining low-latency responsiveness. We introduce a chunkwise diffusion forcing strategy, complemented by a novel self-forcing variant, to alleviate error accumulation and ensure long-term temporal consistency during continuous interaction. By leveraging efficient distillation and system-level optimizations, our framework achieves a stable 25fps at 480p resolution with a time-to-first-frame (TTFF) of only around 1.5 seconds. The proposed method provides holistic and fine-grained full-body control, enabling the agent to transition naturally between diverse behavioral states in interactive scenarios. Experimental results demonstrate that FlowAct-R1 achieves exceptional behavioral vividness and perceptual realism, while maintaining robust generalization across diverse character styles.

---

### Matrix as Plan: Structured Logical Reasoning with Feedback-Driven Replanning

Authors: Ke Chen, Jiandian Zeng, Zihao Peng, Guo Li, Guangxue Zhang, Tian Wang
Published: 2026-01-15
Categories: cs.AI, cs.CL
arXiv: https://arxiv.org/abs/2601.10101v2

As knowledge and semantics on the web grow increasingly complex, enhancing Large Language Models (LLMs)' comprehension and reasoning capabilities has become particularly important. Chain-of-Thought (CoT) prompting has been shown to enhance the reasoning capabilities of LLMs. However, it still falls short on logical reasoning tasks that rely on symbolic expressions and strict deductive rules. Neuro-symbolic methods address this gap by enforcing formal correctness through external solvers. Yet these solvers are highly format-sensitive, and small instabilities in model outputs can lead to frequent processing failures. The LLM-driven approaches avoid parsing brittleness, but they lack structured representations and process-level error-correction mechanisms. To further enhance the logical reasoning capabilities of LLMs, we propose MatrixCoT, a structured CoT framework with a matrix-based plan. Specifically, we normalize and type natural language expressions and attach explicit citation fields, and introduce a matrix-based planning method to preserve global relations among steps. The plan thus becomes a verifiable artifact and execution becomes more stable. For verification, we also add a feedback-driven replanning mechanism. Under semantic-equivalence constraints, it identifies omissions and defects, rewrites and compresses the dependency matrix, and produces a more trustworthy final answer. Experiments on five logical-reasoning benchmarks and five LLMs show that, without relying on external solvers, MatrixCoT enhances both the robustness and interpretability of LLMs when tackling complex symbolic reasoning tasks, while maintaining competitive performance.

---

### InfoSculpt: Sculpting the Latent Space for Generalized Category Discovery

Authors: Wenwen Liao, Hang Ruan, Jianbo Yu, Yuansong Wang, Qingchao Jiang, Xiaofeng Yang
Published: 2026-01-15
Categories: cs.CV
arXiv: https://arxiv.org/abs/2601.10098v1

Generalized Category Discovery (GCD) aims to classify instances from both known and novel categories within a large-scale unlabeled dataset, a critical yet challenging task for real-world, open-world applications. However, existing methods often rely on pseudo-labeling, or two-stage clustering, which lack a principled mechanism to explicitly disentangle essential, category-defining signals from instance-specific noise. In this paper, we address this fundamental limitation by re-framing GCD from an information-theoretic perspective, grounded in the Information Bottleneck (IB) principle. We introduce InfoSculpt, a novel framework that systematically sculpts the representation space by minimizing a dual Conditional Mutual Information (CMI) objective. InfoSculpt uniquely combines a Category-Level CMI on labeled data to learn compact and discriminative representations for known classes, and a complementary Instance-Level CMI on all data to distill invariant features by compressing augmentation-induced noise. These two objectives work synergistically at different scales to produce a disentangled and robust latent space where categorical information is preserved while noisy, instance-specific details are discarded. Extensive experiments on 8 benchmarks demonstrate that InfoSculpt validating the effectiveness of our information-theoretic approach.

---

### Multilingual-To-Multimodal (M2M): Unlocking New Languages with Monolingual Text

Authors: Piyush Singh Pasi
Published: 2026-01-15
Categories: cs.LG
arXiv: https://arxiv.org/abs/2601.10096v2

Multimodal models excel in English, supported by abundant image-text and audio-text data, but performance drops sharply for other languages due to limited multilingual multimodal resources. Existing solutions rely on machine translation, while advances in multilingual text modeling remain underutilized. We introduce M2M, a lightweight alignment method that learns only a few linear layers--using English text alone--to map multilingual text embeddings into multimodal space. Despite its simplicity, M2M matches baseline performance in English (94.9% Recall@10) and achieves strong zero-shot transfer (89.5% Recall@10 averaged across 11 languages, 10 unseen) on XTD Text-to-Image retrieval. Qualitative t-SNE visualizations show that multilingual embeddings align tightly with multimodal representations, while weight analysis reveals that the transformation reshapes embedding geometry rather than performing trivial rotations. Beyond image-text retrieval, M2M demonstrates robustness across datasets and tasks, extending to Audio-Text retrieval and Text-to-Image generation. We release code and checkpoints (https://github.com/piyushsinghpasi/M2M) along with multilingual evaluation datasets: MSCOCO Multilingual 30K (https://huggingface.co/datasets/piyushsinghpasi/mscoco-multilingual-30k), AudioCaps Multilingual (https://huggingface.co/datasets/piyushsinghpasi/audiocaps-multilingual), and Clotho Multilingual (https://huggingface.co/datasets/piyushsinghpasi/clotho-multilingual).

---

### V-Zero: Self-Improving Multimodal Reasoning with Zero Annotation

Authors: Han Wang, Yi Yang, Jingyuan Hu, Minfeng Zhu, Wei Chen
Published: 2026-01-15
Categories: cs.CV, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2601.10094v1

Recent advances in multimodal learning have significantly enhanced the reasoning capabilities of vision-language models (VLMs). However, state-of-the-art approaches rely heavily on large-scale human-annotated datasets, which are costly and time-consuming to acquire. To overcome this limitation, we introduce V-Zero, a general post-training framework that facilitates self-improvement using exclusively unlabeled images. V-Zero establishes a co-evolutionary loop by instantiating two distinct roles: a Questioner and a Solver. The Questioner learns to synthesize high-quality, challenging questions by leveraging a dual-track reasoning reward that contrasts intuitive guesses with reasoned results. The Solver is optimized using pseudo-labels derived from majority voting over its own sampled responses. Both roles are trained iteratively via Group Relative Policy Optimization (GRPO), driving a cycle of mutual enhancement. Remarkably, without a single human annotation, V-Zero achieves consistent performance gains on Qwen2.5-VL-7B-Instruct, improving visual mathematical reasoning by +1.7 and general vision-centric by +2.6, demonstrating the potential of self-improvement in multimodal systems. Code is available at https://github.com/SatonoDia/V-Zero

---

### LeMoF: Level-guided Multimodal Fusion for Heterogeneous Clinical Data

Authors: Jongseok Kim, Seongae Kang, Jonghwan Shin, Yuhan Lee, Ohyun Jo
Published: 2026-01-15
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2601.10092v1

Multimodal clinical prediction is widely used to integrate heterogeneous data such as Electronic Health Records (EHR) and biosignals. However, existing methods tend to rely on static modality integration schemes and simple fusion strategies. As a result, they fail to fully exploit modality-specific representations. In this paper, we propose Level-guided Modal Fusion (LeMoF), a novel framework that selectively integrates level-guided representations within each modality. Each level refers to a representation extracted from a different layer of the encoder. LeMoF explicitly separates and learns global modality-level predictions from level-specific discriminative representations. This design enables LeMoF to achieve a balanced performance between prediction stability and discriminative capability even in heterogeneous clinical environments. Experiments on length of stay prediction using Intensive Care Unit (ICU) data demonstrate that LeMoF consistently outperforms existing state-of-the-art multimodal fusion techniques across various encoder configurations. We also confirmed that level-wise integration is a key factor in achieving robust predictive performance across various clinical conditions.

---