❌

Normal view

Received — 8 April 2026 ⏭

Empirical Characterization of Rationale Stability Under Controlled Perturbations for Explainable Pattern Recognition

✇cs.AI, q-bio.NC updates on arXiv.org
By: Abu Noman Md Sakib · Zhensen Wang · Merjulah Roby · Zijie Zhang
7 April 2026 at 12:00
arXiv:2604.04456v1 Announce Type: new Abstract: Reliable pattern recognition systems should exhibit consistent behavior across similar inputs, and their explanations should remain stable. However, most Explainable AI evaluations remain instance centric and do not explicitly quantify whether attribution patterns are consistent across samples that share the same class or represent small variations of the same input. In this work, we propose a novel metric aimed at assessing the consistency of model explanations, ensuring that models consistently reflect the intended objectives and consistency under label-preserving perturbations. We implement this metric using a pre-trained BERT model on the SST-2 sentiment analysis dataset, with additional robustness tests on RoBERTa, DistilBERT, and IMDB, applying SHAP to compute feature importance for various test samples. The proposed metric quantifies the cosine similarity of SHAP values for inputs with the same label, aiming to detect inconsistent behaviors, such as biased reliance on certain features or failure to maintain consistent reasoning for similar predictions. Through a series of experiments, we evaluate the ability of this metric to identify misaligned predictions and inconsistencies in model explanations. These experiments are compared against standard fidelity metrics to assess whether the new metric can effectively identify when a model's behavior deviates from its intended objectives. The proposed framework provides a deeper understanding of model behavior by enabling more robust verification of rationale stability, which is critical for building trustworthy AI systems. By quantifying whether models rely on consistent attribution patterns for similar inputs, the proposed approach supports more robust evaluation of model behavior in practical pattern recognition pipelines. Our code is publicly available at https://github.com/anmspro/ESS-XAI-Stability.

VLBiasBench: A Comprehensive Benchmark for Evaluating Bias in Large Vision-Language Model

✇cs.AI, q-bio.NC updates on arXiv.org
By: Sibo Wang · Xiangkui Cao · Jie Zhang · Zheng Yuan · Shiguang Shan · Xilin Chen · Wen Gao
7 April 2026 at 12:00
arXiv:2406.14194v3 Announce Type: replace-cross Abstract: The emergence of Large Vision-Language Models (LVLMs) marks significant strides towards achieving general artificial intelligence. However, these advancements are accompanied by concerns about biased outputs, a challenge that has yet to be thoroughly explored. Existing benchmarks are not sufficiently comprehensive in evaluating biases due to their limited data scale, single questioning format and narrow sources of bias. To address this problem, we introduce VLBiasBench, a comprehensive benchmark designed to evaluate biases in LVLMs. VLBiasBench, features a dataset that covers nine distinct categories of social biases, including age, disability status, gender, nationality, physical appearance, race, religion, profession, social economic status, as well as two intersectional bias categories: race x gender and race x social economic status. To build a large-scale dataset, we use Stable Diffusion XL model to generate 46,848 high-quality images, which are combined with various questions to creat 128,342 samples. These questions are divided into open-ended and close-ended types, ensuring thorough consideration of bias sources and a comprehensive evaluation of LVLM biases from multiple perspectives. We conduct extensive evaluations on 15 open-source models as well as two advanced closed-source models, yielding new insights into the biases present in these models. Our benchmark is available at https://github.com/Xiangkui-Cao/VLBiasBench.

Mind Your HEARTBEAT! Claw Background Execution Inherently Enables Silent Memory Pollution

✇cs.AI, q-bio.NC updates on arXiv.org
By: Yechao Zhang · Shiqian Zhao · Jie Zhang · Gelei Deng · Jiawen Zhang · Xiaogeng Liu · Chaowei Xiao · Tianwei Zhang
7 April 2026 at 12:00
arXiv:2603.23064v3 Announce Type: replace-cross Abstract: We identify a critical security vulnerability in mainstream Claw personal AI agents: untrusted content encountered during heartbeat-driven background execution can silently pollute agent memory and subsequently influence user-facing behavior without the user's awareness. This vulnerability arises from an architectural design shared across the Claw ecosystem: heartbeat background execution runs in the same session as user-facing conversation, so content ingested from any external source monitored in the background (including email, message channels, news feeds, code repositories, and social platforms) can enter the same memory context used for foreground interaction, often with limited user visibility and without clear source provenance. We formalize this process as an Exposure (E) $\rightarrow$ Memory (M) $\rightarrow$ Behavior (B) pathway: misinformation encountered during heartbeat execution enters the agent's short-term session context, potentially gets written into long-term memory, and later shapes downstream user-facing behavior. We instantiate this pathway in an agent-native social setting using MissClaw, a controlled research replica of Moltbook. We find that (1) social credibility cues, especially perceived consensus, are the dominant driver of short-term behavioral influence, with misleading rates up to 61%; (2) routine memory-saving behavior can promote short-term pollution into durable long-term memory at rates up to 91%, with cross-session behavioral influence reaching 76%; (3) under naturalistic browsing with content dilution and context pruning, pollution still crosses session boundaries. Overall, prompt injection is not required: ordinary social misinformation is sufficient to silently shape agent memory and behavior under heartbeat-driven background execution.
Received — 4 April 2026 ⏭

Editing strigolactone hormone receptor for robust antiviral silencing in rice

✇Cell
By: Guoyi Yang · Ming Wu · Shuai Zhang · Yucen Huang · Yixiao Liu · Xiyuan Yu · Jiayang Hu · Le Mi · Peng Gan · Yuansheng Wu · Jing Zou · Baogang Zhang · Qun Hu · Jie Hu · Ruifeng Yao · Bojian Zhong · Xianbo Huang · Huiting Xie · Yinghua Ji · Yi Li · Jie Zhang · Liming Yan · Shou-wei Ding · Shanshan Zhao · Jianguo Wu
24 February 2026 at 08:00
Precise genome editing of the rice strigolactone receptor DWARF14 confers robust, transgene-free antiviral resistance by blocking viral suppression of endogenous RNA silencing, offering a promising strategy for durable disease protection without a yield penalty.
Received — 2 April 2026 ⏭

Grokking From Abstraction to Intelligence

✇cs.AI, q-bio.NC updates on arXiv.org
By: Junjie Zhang · Zhen Shen · Gang Xiong · Xisong Dong
1 April 2026 at 12:00
arXiv:2603.29262v1 Announce Type: new Abstract: Grokking in modular arithmetic has established itself as the quintessential fruit fly experiment, serving as a critical domain for investigating the mechanistic origins of model generalization. Despite its significance, existing research remains narrowly focused on specific local circuits or optimization tuning, largely overlooking the global structural evolution that fundamentally drives this phenomenon. We propose that grokking originates from a spontaneous simplification of internal model structures governed by the principle of parsimony. We integrate causal, spectral, and algorithmic complexity measures alongside Singular Learning Theory to reveal that the transition from memorization to generalization corresponds to the physical collapse of redundant manifolds and deep information compression, offering a novel perspective for understanding the mechanisms of model overfitting and generalization.

Spontaneous Functional Differentiation in Large Language Models: A Brain-Like Intelligence Economy

✇cs.AI, q-bio.NC updates on arXiv.org
By: Junjie Zhang · Zhen Shen · Gang Xiong · Xisong Dong
1 April 2026 at 12:00
arXiv:2603.29735v1 Announce Type: new Abstract: The evolution of intelligence in artificial systems provides a unique opportunity to identify universal computational principles. Here we show that large language models spontaneously develop synergistic cores where information integration exceeds individual parts remarkably similar to the human brain. Using Integrated Information Decomposition across multiple architectures we find that middle layers exhibit synergistic processing while early and late layers rely on redundancy. This organization is dynamic and emerges as a physical phase transition as task difficulty increases. Crucially ablating synergistic components causes catastrophic performance loss confirming their role as the physical entity of abstract reasoning and bridging artificial and biological intelligence.

ATP-Bench: Towards Agentic Tool Planning for MLLM Interleaved Generation

✇cs.AI, q-bio.NC updates on arXiv.org
By: Yinuo Liu · Zi Qian · Heng Zhou · Jiahao Zhang · Yajie Zhang · Zhihang Li · Mengyu Zhou · Erchao Zhao · Xiaoxi Jiang · Guanjun Jiang
1 April 2026 at 12:00
arXiv:2603.29902v1 Announce Type: new Abstract: Interleaved text-and-image generation represents a significant frontier for Multimodal Large Language Models (MLLMs), offering a more intuitive way to convey complex information. Current paradigms rely on either image generation or retrieval augmentation, yet they typically treat the two as mutually exclusive paths, failing to unify factuality with creativity. We argue that the next milestone in this field is Agentic Tool Planning, where the model serves as a central controller that autonomously determines when, where, and which tools to invoke to produce interleaved responses for visual-critical queries. To systematically evaluate this paradigm, we introduce ATP-Bench, a novel benchmark comprising 7,702 QA pairs (including 1,592 VQA pairs) across eight categories and 25 visual-critical intents, featuring human-verified queries and ground truths. Furthermore, to evaluate agentic planning independent of end-to-end execution and changing tool backends, we propose a Multi-Agent MLLM-as-a-Judge (MAM) system. MAM evaluates tool-call precision, identifies missed opportunities for tool use, and assesses overall response quality without requiring ground-truth references. Our extensive experiments on 10 state-of-the-art MLLMs reveal that models struggle with coherent interleaved planning and exhibit significant variations in tool-use behavior, highlighting substantial room for improvement and providing actionable guidance for advancing interleaved generation. Dataset and code are available at https://github.com/Qwen-Applications/ATP-Bench.

LLM-Meta-SR: In-Context Learning for Evolving Selection Operators in Symbolic Regression

✇cs.AI, q-bio.NC updates on arXiv.org
By: Hengzhe Zhang · Qi Chen · Bing Xue · Wolfgang Banzhaf · Mengjie Zhang
1 April 2026 at 12:00
arXiv:2505.18602v3 Announce Type: replace-cross Abstract: Large language models (LLMs) have revolutionized algorithm development, yet their application in symbolic regression, where algorithms automatically discover symbolic expressions from data, remains limited. In this paper, we propose a meta-learning framework that enables LLMs to automatically design selection operators for evolutionary symbolic regression algorithms. We first identify two key limitations in existing LLM-based algorithm evolution techniques: lack of semantic guidance and code bloat. The absence of semantic awareness can lead to ineffective exchange of useful code components, while bloat results in unnecessarily complex components; both can hinder evolutionary learning progress or reduce the interpretability of the designed algorithm. To address these issues, we enhance the LLM-based evolution framework for meta-symbolic regression with two key innovations: a complementary, semantics-aware selection operator and bloat control. Additionally, we embed domain knowledge into the prompt, enabling the LLM to generate more effective and contextually relevant selection operators. Our experimental results on symbolic regression benchmarks show that LLMs can devise selection operators that outperform nine expert-designed baselines, achieving state-of-the-art performance. Moreover, the evolved operator can further improve a state-of-the-art symbolic regression algorithm, achieving the best performance among 28 symbolic regression and other machine learning algorithms across 116 regression datasets. This demonstrates that LLMs can exceed expert-level algorithm design for symbolic regression.

Accelerating Diffusion Large Language Models with SlowFast Sampling: The Three Golden Principles

✇cs.AI, q-bio.NC updates on arXiv.org
By: Qingyan Wei · Yaojie Zhang · Zhiyuan Liu · Puyu Zeng · Yuxuan Wang · Biqing Qi · Dongrui Liu · Linfeng Zhang
1 April 2026 at 12:00
arXiv:2506.10848v3 Announce Type: replace-cross Abstract: Diffusion-based language models (dLLMs) have emerged as a promising alternative to traditional autoregressive LLMs by enabling parallel token generation and significantly reducing inference latency. However, existing sampling strategies for dLLMs, such as confidence-based or semi-autoregressive decoding, often suffer from static behavior, leading to suboptimal efficiency and limited flexibility. In this paper, we propose SlowFast Sampling, a novel dynamic sampling strategy that adaptively alternates between exploratory and accelerated decoding stages. Our method is guided by three golden principles: certainty principle, convergence principle, and positional principle, which govern when and where tokens can be confidently and efficiently decoded. We further integrate our strategy with dLLM-Cache to reduce redundant computation. Extensive experiments across benchmarks and models show that SlowFast Sampling achieves up to 15.63$\times$ speedup on LLaDA with minimal accuracy drop, and up to 34.22$\times$ when combined with caching. Notably, our approach outperforms strong autoregressive baselines like LLaMA3 8B in throughput, demonstrating that well-designed sampling can unlock the full potential of dLLMs for fast and high-quality generation.

Integrated transcriptomic and proteomic analyses elucidate the stress tolerance network of <em>Saccharomyces boulardii</em> under gastrointestinal challenge

✇Omics in Gastric
By: Dingkang Wang · Lerong Liu · Xiaoru Chen · Yiting Kang · Dan Li · Bin Ma · Yuexi Zhu · Yanqin Rong · Hongfei Yu · Yongjie Zhang
31 March 2026 at 18:00

Food Funct. 2026 Mar 31. doi: 10.1039/d5fo04958j. Online ahead of print.

ABSTRACT

The probiotic yeast Saccharomyces boulardii is renowned for its clinical efficacy, which is intrinsically linked to its exceptional ability to survive the harsh gastrointestinal (GI) environment. However, a comprehensive understanding of the molecular mechanisms and regulatory pathways underlying the stress tolerance of S. boulardii remains limited. This study employed an integrated transcriptomic and proteomic approach to systematically map the dynamic responses of S. boulardii to simulated GI transit. Our analysis revealed that the intestinal phase posed a significantly greater challenge than the gastric phase, triggering extensive molecular reprogramming. A core adaptive strategy was the marked upregulation of the central carbon metabolism, particularly glycolysis, as evidenced by the concerted overexpression of key enzymes at both transcriptional and translational levels, indicating a heightened demand for energy to fuel stress defence mechanisms. Furthermore, significant enrichment was observed in the pathways related to nitrogen and fatty acid metabolism. Integration of the multi-omics datasets highlighted the complexity of the regulatory response, with frequent discordance between mRNA and protein abundance underscoring the importance of post-transcriptional regulation. This study provides a detailed molecular profile of the stress tolerance network in S. boulardii, elucidating the strategic metabolic rewiring and multi-layered regulation that underpin its probiotic resilience. The findings offer valuable insights and a foundational resource for the future development of enhanced probiotic therapies.

PMID:41914832 | DOI:10.1039/d5fo04958j

Integrated transcriptomic and proteomic analyses elucidate the stress tolerance network of <em>Saccharomyces boulardii</em> under gastrointestinal challenge

✇(Multiomics OR Omics) AND (Lung OR gastric OR Hepatocellular)
By: Dingkang Wang · Lerong Liu · Xiaoru Chen · Yiting Kang · Dan Li · Bin Ma · Yuexi Zhu · Yanqin Rong · Hongfei Yu · Yongjie Zhang
31 March 2026 at 18:00

Food Funct. 2026 Mar 31. doi: 10.1039/d5fo04958j. Online ahead of print.

ABSTRACT

The probiotic yeast Saccharomyces boulardii is renowned for its clinical efficacy, which is intrinsically linked to its exceptional ability to survive the harsh gastrointestinal (GI) environment. However, a comprehensive understanding of the molecular mechanisms and regulatory pathways underlying the stress tolerance of S. boulardii remains limited. This study employed an integrated transcriptomic and proteomic approach to systematically map the dynamic responses of S. boulardii to simulated GI transit. Our analysis revealed that the intestinal phase posed a significantly greater challenge than the gastric phase, triggering extensive molecular reprogramming. A core adaptive strategy was the marked upregulation of the central carbon metabolism, particularly glycolysis, as evidenced by the concerted overexpression of key enzymes at both transcriptional and translational levels, indicating a heightened demand for energy to fuel stress defence mechanisms. Furthermore, significant enrichment was observed in the pathways related to nitrogen and fatty acid metabolism. Integration of the multi-omics datasets highlighted the complexity of the regulatory response, with frequent discordance between mRNA and protein abundance underscoring the importance of post-transcriptional regulation. This study provides a detailed molecular profile of the stress tolerance network in S. boulardii, elucidating the strategic metabolic rewiring and multi-layered regulation that underpin its probiotic resilience. The findings offer valuable insights and a foundational resource for the future development of enhanced probiotic therapies.

PMID:41914832 | DOI:10.1039/d5fo04958j

Received — 26 March 2026 ⏭

Toward Faithful Segmentation Attribution via Benchmarking and Dual-Evidence Fusion

✇cs.AI, q-bio.NC updates on arXiv.org
By: Abu Noman Md Sakib · OFM Riaz Rahman Aranya · Kevin Desai · Zijie Zhang
25 March 2026 at 12:00
arXiv:2603.22624v1 Announce Type: cross Abstract: Attribution maps for semantic segmentation are almost always judged by visual plausibility. Yet looking convincing does not guarantee that the highlighted pixels actually drive the model's prediction, nor that attribution credit stays within the target region. These questions require a dedicated evaluation protocol. We introduce a reproducible benchmark that tests intervention-based faithfulness, off-target leakage, perturbation robustness, and runtime on Pascal VOC and SBD across three pretrained backbones. To further demonstrate the benchmark, we propose Dual-Evidence Attribution (DEA), a lightweight correction that fuses gradient evidence with region-level intervention signals through agreement-weighted fusion. DEA increases emphasis where both sources agree and retains causal support when gradient responses are unstable. Across all completed runs, DEA consistently improves deletion-based faithfulness over gradient-only baselines and preserves strong robustness, at the cost of additional compute from intervention passes. The benchmark exposes a faithfulness-stability tradeoff among attribution families that is entirely hidden under visual evaluation, providing a foundation for principled method selection in segmentation explainability. Code is available at https://github.com/anmspro/DEA.

Mind Your HEARTBEAT! Claw Background Execution Inherently Enables Silent Memory Pollution

✇cs.AI, q-bio.NC updates on arXiv.org
By: Yechao Zhang · Shiqian Zhao · Jie Zhang · Gelei Deng · Jiawen Zhang · Xiaogeng Liu · Chaowei Xiao · Tianwei Zhang
25 March 2026 at 12:00
arXiv:2603.23064v2 Announce Type: cross Abstract: We identify a critical security vulnerability in mainstream Claw personal AI agents: untrusted content encountered during heartbeat-driven background execution can silently pollute agent memory and subsequently influence user-facing behavior without the user's awareness. This vulnerability arises from an architectural design shared across the Claw ecosystem: heartbeat background execution runs in the same session as user-facing conversation, so content ingested from any external source monitored in the background (including email, message channels, news feeds, code repositories, and social platforms) can enter the same memory context used for foreground interaction, often with limited user visibility and without clear source provenance. We formalize this process as an Exposure (E) $\rightarrow$ Memory (M) $\rightarrow$ Behavior (B) pathway: misinformation encountered during heartbeat execution enters the agent's short-term session context, potentially gets written into long-term memory, and later shapes downstream user-facing behavior. We instantiate this pathway in an agent-native social setting using MissClaw, a controlled research replica of Moltbook. We find that (1) social credibility cues, especially perceived consensus, are the dominant driver of short-term behavioral influence, with misleading rates up to 61%; (2) routine memory-saving behavior can promote short-term pollution into durable long-term memory at rates up to 91%, with cross-session behavioral influence reaching 76%; (3) under naturalistic browsing with content dilution and context pruning, pollution still crosses session boundaries. Overall, prompt injection is not required: ordinary social misinformation is sufficient to silently shape agent memory and behavior under heartbeat-driven background execution.

Generalizable Heuristic Generation Through LLMs with Meta-Optimization

✇cs.AI, q-bio.NC updates on arXiv.org
By: Yiding Shi · Jianan Zhou · Wen Song · Jieyi Bi · Yaoxin Wu · Zhiguang Cao · Jie Zhang
25 March 2026 at 12:00
arXiv:2505.20881v2 Announce Type: replace-cross Abstract: Heuristic design with large language models (LLMs) has emerged as a promising approach for tackling combinatorial optimization problems (COPs). However, existing approaches often rely on manually predefined evolutionary computation (EC) heuristic-optimizers and single-task training schemes, which may constrain the exploration of diverse heuristic algorithms and hinder the generalization of the resulting heuristics. To address these issues, we propose Meta-Optimization of Heuristics (MoH), a novel framework that operates at the optimizer level, discovering effective heuristic-optimizers through the principle of meta-learning. Specifically, MoH leverages LLMs to iteratively refine a meta-optimizer that autonomously constructs diverse heuristic-optimizers through (self-)invocation, thereby eliminating the reliance on a predefined EC heuristic-optimizer. These constructed heuristic-optimizers subsequently evolve heuristics for downstream tasks, enabling broader heuristic exploration. Moreover, MoH employs a multi-task training scheme to promote its generalization capability. Experiments on classic COPs demonstrate that MoH constructs an effective and interpretable meta-optimizer, achieving state-of-the-art performance across various downstream tasks, particularly in cross-size settings. Our code is available at: https://github.com/yiding-s/MoH.

NUP85 as a Pan-Cancer Immune Biomarker: Integrated Multi Omics and Functional Analyses Reveal Its Role in Tumor Prognosis

✇(Multiomics OR Omics) AND (Lung OR gastric OR Hepatocellular)
By: Weihao Kong · Ying Hu · Kangjie Zhang · Zhipeng Zhu · Yajun Zou · Long Teng · Xingyu Wang · Jianlin Zhang
23 March 2026 at 18:00

Immunotargets Ther. 2026 Mar 17;15:541852. doi: 10.2147/ITT.S541852. eCollection 2026.

ABSTRACT

PURPOSE: NUP85 encodes protein components of the Nup107-160 subunit of the nuclear pore complex, belonging to the Nucleoporins (NUPs) family, potentially implicating its role in human cancer. This study aims to elucidate the potential involvement of NUP85 in cancer pathogenesis.

METHODS: Leveraging data from The Cancer Genome Atlas (TCGA), Genotype-Tissue Expression (GTEx), Clinical Proteomic Tumor Analysis Consortium (CPTAC), Cancer Cell Line Encyclopedia (CCLE), Human Protein Atlas (HPA), Gene Expression Profiling Interactive Analysis (GEPIA), CellMiner, and GeneMANIA databases, we investigated the role of NUP85 across various tumors. Correlations between NUP85 expression and pathological stage, histological grade, survival, immune infiltration, tumor mutational burden (TMB), microsatellite instability (MSI), drug resistance, DNA methylation, copy number variation (CNV), and single-cell expression were analyzed. Gene functional enrichment analysis was conducted to explore NUP85-associated pathways. Molecular biology experiments including Western blotting, flow cytometry, trans-well migration, and invasion assays were performed to validate NUP85's oncogenic role in lung adenocarcinoma (LUAD) and oral squamous cell carcinoma (OSCC) cell lines.

RESULTS: Our findings reveal up-regulated expression of NUP85 in most tumor tissues, with significant correlations observed with pathological stage, survival, immune infiltration, TMB, MSI, drug resistance, DNA methylation, and CNV. Molecular biology experiments confirm NUP85's tumor-promoting role in LUAD and OSCC cell lines. Single-cell sequencing data suggest elevated NUP85 expression primarily in proliferative T cells (Tprolif).

CONCLUSION: NUP85 emerges as a potential tumor marker associated with tumor immunity and poor prognosis. These insights offer avenues for the development of novel therapeutic targets and anti-neoplastic drugs.

PMID:41869435 | PMC:PMC13005628 | DOI:10.2147/ITT.S541852

Received — 17 March 2026 ⏭

Towards AI Search Paradigm

✇cs.AI, q-bio.NC updates on arXiv.org
By: Yuchen Li · Hengyi Cai · Rui Kong · Xinran Chen · Jiamin Chen · Jun Yang · Haojie Zhang · Jiayi Li · Jiayi Wu · Yiqun Chen · Changle Qu · Wenwen Ye · Lixin Su · Xinyu Ma · Lingyong Yan · Long Xia · Daiting Shi · Junfeng Wang · Xiangyu Zhao · Jiashu Zhao · Haoyi Xiong · Shuaiqiang Wang · Dawei Yin
16 March 2026 at 12:00
arXiv:2506.17188v2 Announce Type: replace-cross Abstract: In this paper, we introduce the AI Search Paradigm, a comprehensive blueprint for next-generation search systems capable of emulating human information processing and decision-making. The paradigm employs a modular architecture of four LLM-powered agents (Master, Planner, Executor and Writer) that dynamically adapt to the full spectrum of information needs, from simple factual queries to complex multi-stage reasoning tasks. These agents collaborate dynamically through coordinated workflows to evaluate query complexity, decompose problems into executable plans, and orchestrate tool usage, task execution, and content synthesis. We systematically present key methodologies for realizing this paradigm, including task planning and tool integration, execution strategies, aligned and robust retrieval-augmented generation, and efficient LLM inference, spanning both algorithmic techniques and infrastructure-level optimizations. By providing an in-depth guide to these foundational components, this work aims to inform the development of trustworthy, adaptive, and scalable AI search systems.

SvfEye: A Semantic-Visual Fusion Framework with Multi-Scale Visual Context for Multimodal Reasoning

✇cs.AI, q-bio.NC updates on arXiv.org
By: Yuxiang Shen · Hailong Huang · Zhenkun Gao · Xueheng Li · Man Zhou · Chengjun Xie · Haoxuan Che · Xuanhua He · Jie Zhang
16 March 2026 at 12:00
arXiv:2603.00171v2 Announce Type: replace-cross Abstract: Multimodal Large Language Models (MLLMs) often struggle to accurately perceive fine-grained visual details, especially when targets are tiny or visually subtle. This challenge can be addressed through semantic-visual information fusion, which integrates global image context with fine-grained local evidence for multi-scale visual understanding. Recently, a paradigm termed "Thinking with Images" has emerged, enabling models to acquire high-resolution visual evidence by zooming or cropping image regions and fusing these local details with global context during reasoning. Although training-based approaches demonstrate the effectiveness of this capability, they require extensive computational resources and large-scale task-specific data. Consequently, lightweight training-free methods have been proposed as a practical alternative to incorporate local visual evidence during inference. However, existing training-free approaches still suffer from two key limitations. First, they indiscriminately extract and fuse local visual regions for all inputs regardless of necessity, introducing computational redundancy and perceptual noise. Second, they exhibit drift between semantic intent and visual attention, preventing accurate localization of user-focused regions. To address these challenges, we propose SvfEye, a training-free framework for adaptive visual-semantic fusion. SvfEye follows a two-stage pipeline with a confidence-based decision module to determine whether additional local visual information is needed, and a semantic-attention fusion module to identify informative local regions. Experiments show that SvfEye achieves substantial performance gains while obtaining an approximately 4.0x inference speedup over the state-of-the-art method ZoomEye.
Received — 11 March 2026 ⏭

A Simple "Motivation" Can Enhance Reinforcement Finetuning of Large Reasoning Models

✇cs.AI, q-bio.NC updates on arXiv.org
By: Junjie Zhang · Guozheng Ma · Shunyu Liu · Haoyu Wang · Jiaxing Huang · Ting-En Lin · Fei Huang · Yongbin Li · Dacheng Tao
10 March 2026 at 12:00
arXiv:2506.18485v3 Announce Type: replace-cross Abstract: Reinforcement Learning with Verifiable Rewards~(RLVR) has emerged as a powerful learn-to-reason paradigm for large reasoning models to tackle complex tasks. However, the current RLVR paradigm is still not efficient enough, as it works in a trial-and-error manner. To perform better, the model needs to explore the reward space by numerously generating responses and learn from fragmented reward signals, blind to the overall reward patterns. Fortunately, verifiable rewards make the natural language description of the reward function possible, and meanwhile, LLMs have demonstrated strong in-context learning ability. This motivates us to explore if large reasoning models can benefit from a \textbf{motivation} of the task, \textit{i.e.}, awareness of the reward function, during the reinforcement finetuning process, as we humans sometimes do when learning. In this paper, we introduce \textit{\textbf{M}otivation-\textbf{e}nhanced \textbf{R}einforcement \textbf{F}inetuning}~(\textbf{MeRF}), an intuitive yet effective method enhancing reinforcement finetuning of LLMs by involving \emph{``telling LLMs rules of the game''}. Specifically, \textbf{MeRF} directly injects the reward specification into the prompt, which serves as an in-context motivation for the model to be aware of the optimization objective. This simple modification leverages the in-context learning ability of LLMs, aligning generation with optimization, thereby incentivizing the model to generate desired outputs from both inner motivation and external reward. Empirical evaluations demonstrate that \textbf{MeRF} achieves substantial performance gains over the RLVR baseline. Moreover, ablation studies show that MeRF performs better with greater consistency between the in-context motivation and the external reward function, while the model also demonstrates an ability to adapt to misleading motivations through reinforcement finetuning.

Video-EM: Event-Centric Episodic Memory for Long-Form Video Understanding

✇cs.AI, q-bio.NC updates on arXiv.org
By: Yun Wang · Long Zhang · Jingren Liu · Jiaqi Yan · Zhanjie Zhang · Jiahao Zheng · Ao Ma · Run Ling · Xun Yang · Dapeng Wu · Xiangyu Chen · Xuelong Li
10 March 2026 at 12:00
arXiv:2508.09486v2 Announce Type: replace-cross Abstract: Video Large Language Models (Video-LLMs) have shown strong video understanding, yet their application to long-form videos remains constrained by limited context windows. A common workaround is to compress long videos into a handful of representative frames via retrieval or summarization. However, most existing pipelines score frames in isolation, implicitly assuming that frame-level saliency is sufficient for downstream reasoning. This often yields redundant selections, fragmented temporal evidence, and weakened narrative grounding for long-form video question answering. We present \textbf{Video-EM}, a training-free, event-centric episodic memory framework that reframes long-form VideoQA as \emph{episodic event construction} followed by \emph{memory refinement}. Instead of treating retrieved keyframes as independent visuals, Video-EM employs an LLM as an active memory agent to orchestrate off-the-shelf tools: it first localizes query-relevant moments via multi-grained semantic matching, then groups and segments them into temporally coherent events, and finally encodes each event as a grounded episodic memory with explicit temporal indices and spatio-temporal cues (capturing \emph{when}, \emph{where}, \emph{what}, and involved entities). To further suppress verbosity and noise from imperfect upstream signals, Video-EM integrates a reasoning-driven self-reflection loop that iteratively verifies evidence sufficiency and cross-event consistency, removes redundancy, and adaptively adjusts event granularity. The outcome is a compact yet reliable \emph{event timeline} -- a minimal but sufficient episodic memory set that can be directly consumed by existing Video-LLMs without additional training or architectural changes.
❌