DynaKRAG is a unified framework that learns a state‑conditioned policy to control evidence acquisition in multi‑hop retrieval‑augmented generation. It uses a deterministic validity layer to build an action set, a learned continuation gate to decide between generating an answer or gathering more evidence, and an advantage scorer to rank evidence operations by predicted gain. Across HotpotQA, 2Wiki, and MuSiQue with various backbone models, DynaKRAG achieves top EM and F1 scores, improves token and retrieval efficiency, and enables terminal evidence compression that reduces context size while boosting answer quality.
By Chenyu Zhou, Yaqi Wu, Xiaolei Guo, Jiaqi Huang, Xianfa Zhang, Junxu Zhang, Zhuo Yu, Zhubo Shi, Jianghao Lin, Dongdong Ge
arXiv:2608.23503v1 Announce Type: new
Abstract: Text-based person anomaly search requires distinguishing individuals based on fine-grained, context-dependent behaviors rather than mere appearance. Ex...
By Thanh-Khoi Nguyen, Thanh-Nhan Vo, Trong-Thuan Nguyen, Minh-Triet Tran
arXiv:2608.21367v1 Announce Type: cross
Abstract: Protein-peptide interactions are central to cellular regulation and peptide-based drug discovery, yet existing computational methods mainly focus on...
By Hao Qian, Shikui Tu, Lei Xu
The paper introduces the concept of LLM‑specific utility, defining it as the performance gain a target large language model (LLM) achieves when provided with a passage compared to answering without evidence. A benchmark of utilitarian passages is built for four LLMs (Qwen3‑8B/14B/32B and Llama 3.1‑8B) across three QA datasets, revealing that each model benefits most from its own tailored evidence and that evidence optimized for other models is consistently suboptimal. The authors also create SpecUBench, a benchmark for LLM‑specific utility judgment, and show that current utility‑aware retrieval methods largely capture model‑agnostic usefulness, struggling to estimate LLM‑specific utility.
"whyItMatters":"The study demonstrates that retrieval‑augmented generation must consider model‑specific evidence selection to truly improve LLM performance, highlighting a gap in existing utility‑aware methods."
By Hengran Zhang, Keping Bi, Jiafeng Guo, Jiaming Zhang, Shuaiqiang Wang, Dawei Yin, Xueqi Cheng
ExtrinSplat is a new framework that separates geometry from semantics in 3D Gaussian Splatting scenes. It clusters Gaussians into overlapping 3D object groups and uses a Vision‑Language Model to generate lightweight textual hypotheses, creating an extrinsic index layer that handles complex polysemy. This approach reduces adaptation time from hours to minutes, cuts storage overhead by orders of magnitude, and outperforms existing embedding‑based methods on open‑vocabulary 3D object selection and semantic segmentation benchmarks.
By Jiayu Ding, Xinpeng Liu, Zhiyi Pan, Shiqiang Long, Ge Li
arXiv:2608.21464v1 Announce Type: cross
Abstract: We demonstrate that structured distortion of training data - which we term complexity induction - can induce compositional generalization in a standa...
By Aleksandr Abramov
arXiv:2608.22066v1 Announce Type: cross
Abstract: Attention-based multiple instance learning (ABMIL) using pathology foundation model embeddings is effective for slide-level tasks, but exhaustive inf...
By Duncan Stothers, Ren-Chin Wu, William Lotter
DamageScope is a retrieval‑augmented framework that combines satellite imagery, Vision‑Language Models (VLMs), and Large Language Models (LLMs) to automate property damage assessment after natural disasters. It uses a Retrieval‑Augmented Generation (RAG) architecture to extract structured visual representations from satellite images, enabling interactive natural language queries. The system introduces a multi‑vector embedding‑based clustering algorithm that improves scalability and reduces indexing time by up to 14×, and a dual‑store data architecture that cuts LLM API calls, lowering operational cost and response latency by roughly 3×.
By Ravi K. Rajendran, Biplob Debnath, Murugan Sankaradas, Srimat T. Chakradhar
Although Speech Large Language Models (SpeechLLMs) excel at speech understanding and generation, their capacity for fine-grained, temporally aligned outputs remains underexplored. Our work addresses t...
Large Language Models excel at code generation, yet competitive programming exposes a persistent failure mode: existing multi-agent pipelines distribute work over generic planner, coder, and debugger...
Negative sampling determines whether a knowledge graph embedding (KGE) model learns from informative counterexamples or wastes updates on implausible corruptions. Uniform negatives are diverse but eas...
We present Team Semiintelligencn's solution for the ACM RecSys 2026 TalkPlayData Challenge, addressing conversational music recommendation through a multi-modal and personalized conversational recomme...
The quadratic growth of attention computation and key-value (KV) cache with respect to sequence length is a central bottleneck for ultra-long-context language models and high-resolution generative mod...
The article titled "10 Positions for Enterprise RAG That Mainstream Tutorials Get Wrong" discusses ten key positions related to Enterprise Document Intelligence, as outlined in the series "Enterprise Document Intelligence [Vol.1 #M3]. It also provides a map of every article that supports these positions.
By Kezhan Shi
Neurobiological studies and representation learning have observed that representations of objects belonging to the same category in high-dimensional neural spaces exhibit low-dimensional object manifo...
arXiv:2608. 20810v1 Announce Type: cross Abstract: Multimodal information systems increasingly route generated visual content back through the same vision-language index that informed its production, so the output must remain retrievable by the queries it was meant to serve.
By Guangyuan Dong, Chuang Liu, Yangchen Zeng, Haoyu Wang, Xiaoyang Yu, Pinlong Zhao, Yuchao Hou, Ziwei Li, Zheng Lin
arXiv:2509.20377v2 Announce Type: replace-cross
Abstract: Retrieval-Augmented Generation (RAG) has significantly improved the performance of large language models (LLMs) on knowledge-intensive tasks...
By Tomoaki Isoda
arXiv:2608.21147v1 Announce Type: new
Abstract: The cyclic structure of physiological processes offers a natural prior for self-supervised representation learning, and the cardiac cycle provides a pa...
By Blaise Delaney, Dominic Dootson, Juan Jose Juan Castella, Salil Patel, Andrew Pfaff, Yuji Xing, Jonny Hancox, Karin Sevegnani
SENTRY is a deterministic risk assessment platform for IT change management that replaces subjective questionnaires with a machine‑learning pipeline using XGBoost and retrieval‑augmented generation. It integrates structured metadata, dependency graphs, incident history, and a hybrid semantic‑lexical search of change request text, compressing the unstructured signal into a single feature for model inference. The system delivers a ROC AUC of 0.87, 85% accuracy, and detects high‑risk changes 3.25 times faster than existing processes.
By Daniel Arulpragasam, Christer Henrysson, Ella Ly, Deepika Anbalagan, Leo Feng
The paper presents an end‑to‑end framework for extracting and clustering trilingual Sri Lankan parliamentary debates in Sinhala, Tamil, and English. Using LLM‑based text extraction, multilingual embeddings, and density‑based clustering, the authors recover 30 macro‑topics with a cluster purity of 0.673. The temporal patterns of these topics align with major national events such as the 2019 Easter attacks and the 2022 economic crisis, demonstrating the method’s effectiveness where traditional LDA fails.
By Himath Dhanapala, Haren Daishika, Himandhi Kuruppu, Sithija Seneviratne, Ashini Kavindya, Patalee Narasinghe, Sandeepa Weerasekara, Nisansa de Silva, Sandareka Wickramanayake