Matryoshka Hash Representations (MHR) propose a two‑stage quantization approach for retrieval‑augmented generation. First, a long binary code is learned; then, frozen, additional zero‑initialized residual adaptors are trained to produce searchable prefixes of varying byte budgets. Evaluated on MS MARCO and transferred to seven BEIR datasets, MHR achieves higher NDCG@10 and Recall@100 at 32‑byte budgets than baselines, especially in low‑budget regimes, and can also improve candidate shortlisting and graph‑index pruning.
By Peichun Hua, Yunming Xiao
arXiv:2609.10008v1 Announce Type: new
Abstract: Composed video retrieval (CoVR) searches a gallery for the target video that realizes a natural-language modification of a source clip. However, at gal...
By Dmitry Demidov, Muhammad Zaigham Zaheer, Omkar Thawakar, Abdelrahman Mohamed Shaker, Rao Anwer
The paper introduces Bio-Memory, a biometric-aware memory architecture for large language model agents that conditions memory retrieval on both semantic similarity and biometric matching. By augmenting each memory note with a biometric embedding, Bio-Memory filters retrieval candidates based on identity before ranking them semantically. Experiments on a 10-user shared-agent setting across face and palmprint benchmarks show significant improvements in distinguishing owner from non-owner queries, with up to a 27.29% F1 gain on CALFW and 25.75% on MS_Blue.
By Yanhong Qian, Qingguo Meng, Shihao Ding, Xingbo Dong, Zhe Jin, Hanrui Wang, Isao Echizen
The paper introduces a quantile‑led feature‑extraction framework for predictive maintenance that separates representation learning from downstream modeling. Using a dual‑stage MLP‑QRNN hierarchy, it learns a ten‑quantile distribution per sensor and refines a mid‑tail quantile set into compact, channel‑resolved features. Experiments on 72 machines across nine facilities show that increasing the retained mid‑tail quantiles improves short‑term F1 scores, and that horizon‑conditioned extractors outperform a fixed short‑horizon extractor, demonstrating that feature extraction should be horizon‑dependent rather than fixed preprocessing.
By David J Poland, Daniele Ravi, Na Helian
EmoMed is a multimodal medical consultation agent that tailors its responses to users' emotional states—such as anxiety, confusion, or urgency—while preserving clinical accuracy. It processes text and medical images, detects affect indicators, and adjusts tone, structure, and detail accordingly. The system ensures factual reliability through a dual retrieval mechanism that combines web-based fact‑checking with an API‑connected, continuously updated medical knowledge base, and it has been evaluated across seven state‑of‑the‑art language models using comprehensive metrics, showing that emotionally adaptive responses outperform neutral baselines without sacrificing accuracy.
By Ivan Nasonov, Nikita Glazkov, Ivan Makovetskiy, Mikhail Mozikov, Daniil Sukhorukov, Andrey Savchenko, Ilya Makarov
The paper introduces a Symbolic-RAG-Generative architecture called GRACE for goal‑oriented conversational systems. GRACE transforms business intent into a fixed objective set and uses a constrained policy to update the conversation state based on visitor‑authored evidence while ensuring visitor utility. Evaluation on real‑estate and professional‑cleaning dialogues shows high accuracy in state transitions, evidence precision/recall, and monotonicity.
By Ramon Gonzalez (Mentomy AI), Antonio Diaz (Mentomy AI)
The paper introduces Graph‑Agentic Retrieval‑Augmented Generation (RAG), a system that blends structured evidence with adaptive agents capable of planning retrieval, navigating relations, verifying claims, delegating tasks, and employing tools. It highlights how defects in graph construction can propagate through retrieval and control decisions, potentially leading to significant outcomes. To address these risks, the authors propose an assurance‑by‑construction framework with five interface contracts—evidence, retrieval, reasoning, capability & delegation, and outcome—that make provenance, validity, authorization, uncertainty, and recoverability explicit, and outline an evaluation agenda for social‑good applications.
By Vijay Bommireddy, Raviteja Bommireddy
ARC‑Bench is a new benchmark that tests whether frozen JEPA‑style latent world models can correctly rank candidate actions by latent distance. The study finds that the assumption of latent rankability fails dramatically in both navigation and manipulation tasks, with the top‑scored actions often being suboptimal. Closed‑loop replanning masks this defect, but reducing replanning frequency reveals the underlying ranking failures.
By Zhengshu Zhang, Zhiyuan Li
The paper introduces ENOUGH, a method for creating minimal sufficient user profiles for personalized language models. ENOUGH iteratively adds behavioral records or stops, evaluating profile prefixes with a counterfactual search that balances downstream gains, user specificity, and token costs. The resulting profiles are distilled into a lightweight controller that orders records and triggers the generator only once, achieving better effectiveness and efficiency than existing baselines across six tasks.
By Minghang Liu, Qiang Qiu, Yuanzhuo Wang, Huawei Shen, Xueqi Cheng
The paper proposes a new algorithm for generating Semantic IDs that are both unique and preserve the structure of the original embedding space. By employing bottom‑up clustering, the method maintains local structure, leading to higher clustering quality. This improved structure enhances the utility of the Semantic IDs for downstream generative retrieval tasks.
By Leah Woldemariam, Sudhanshu Garg, Taha Belkhouja, Charles Kim-Yip, Ali Sahami
Noesis is a retrieval-augmented architecture designed for small local language models that prioritizes deterministic judgments before generation. It employs a fact layer, positional addressing, provenance scoping, and two-tier context hydration to ensure factual integrity, achieving parity with larger models on exact value accuracy and eliminating confabulated numbers. The system delivers single-generation queries with traceable, source‑linked outputs, improving performance in regulated domains.
By Nicola Cogotti
The paper maps the nascent social‑science literature on large language models (LLMs) by analysing 198 curated papers and 47,719 field‑scale papers. It identifies three main domains—LLM as Social Minds, LLM Societies, and LLM‑Human Interactions—each containing 13 subcategories such as reasoning, bias, collective intelligence, and trust. The taxonomy is validated through clustering stability, author classification agreement, and topic mapping, revealing differing prominence across conference and journal venues.
By Yi Yang, Xiao Jia, Zeyun Dong, Chenzhang Wang, Zhanzhan Zhao
The paper introduces a bounded spectral framework to analyze rotary attention in transformer language models, focusing on phase alignment, hidden‑state continuity, and semantic drift. It identifies ordered hidden‑state sequences as suitable domains for spectral decomposition, derives the Rotary Position Embedding (RoPE) attention score as a sum of magnitude‑weighted cosine terms, and proves a local stability lemma linking bounded phase displacement to pre‑softmax score degradation. By defining complex modal coordinates and a weighted coherence functional, the work distinguishes representational continuity from execution‑boundary admissibility, offering a theoretical program for when spectral structure explains continuity and when external governance is required.
By Abraham Chachamovits
Vague2Detect is a hybrid pipeline that improves object detection for ambiguous prompts by combining a fine‑tuned Sentence‑BERT to retrieve candidates from a structured household knowledge base, YOLO‑World to verify their presence in images, and a GPT‑3.5‑turbo fallback to generate new candidate descriptions when prompts fall outside the knowledge base. On a benchmark of household scenes, Vague2Detect raises the vague prompt success rate from 32% (YOLO‑World alone) to 61% with high precision, and up to 85% when the GPT fallback is used.
By Ibrohimjon Muminov (Dongguk University, Seoul, South Korea), Jihie Kim (Dongguk University, Seoul, South Korea)
The paper investigates how many transformer components influence a token prediction by measuring the absolute contribution of each unit and channel to the logit. It finds that thousands of components contribute to a single prediction, yet a small subset—often just dozens—carries the majority of the predictive mass. Across models ranging from 124 M to 7 B parameters, the proportion of the model involved in a prediction remains around one to three percent, independent of size, and the study demonstrates that specific components can be directly read and written to modify model behavior without additional training.
By Mark Oskin
The paper introduces K-Token Merging, a latent-space compression method that merges each contiguous block of K token embeddings into a single embedding using a lightweight encoder. The compressed sequence is then processed by a LoRA-adapted large language model, while generation continues in the original vocabulary. Experiments on tasks such as structural reasoning, sentiment classification, and code editing demonstrate that K-Token Merging achieves up to 75% input length reduction with minimal performance loss, placing it on the Pareto frontier of performance versus compression.
By Zihao Xu, John Harvill, Ziwei Fan, Yizhou Sun, Hao Ding, Hao Wang
MoEMB introduces a mixture‑of‑experts (MoE) approach to scale universal multimodal embeddings (UME) without increasing the size of the output vector or relying on autoregressive decoding. By expanding encoder capacity along the expert axis, MoEMB achieves state‑of‑the‑art performance on MMEB‑V2 and MRMR benchmarks with only 3 B active parameters, outperforming TTE‑based methods that use more than four times as many active parameters and require significantly more compute. The paper also presents the first comprehensive study of adaptive computation for MoE‑based embeddings, exploring training‑time and inference‑time strategies to further improve efficiency for large‑scale retrieval and recommendation systems.
By Xuanming Cui, Shlok Kumar Mishra, Wentao Bao, Aashu Singh, Zihao Wang, Xiangjun Fan, Jun Xiao, Ser-Nam Lim, Jianpeng Cheng
DC-Gen is a post‑training framework that accelerates text‑to‑image diffusion models by using a deeply compressed latent space. It first aligns the base model’s latent representations with a lightweight embedding alignment, then applies minimal LoRA fine‑tuning to preserve generation quality. Experiments on SANA and FLUX.1‑Krea show that DC‑Gen‑FLUX cuts 4K image generation latency by 53× on an NVIDIA H100 and, with NVFP4 SVDQuant, achieves a 138× total speedup on a single NVIDIA 5090 GPU.
By Wenkun He, Yuchao Gu, Junyu Chen, Dongyun Zou, Yujun Lin, Zhekai Zhang, Haocheng Xi, Muyang Li, Ligeng Zhu, Jincheng Yu, Junsong Chen, Enze Xie, Song Han, Han Cai
arXiv:2609.08173v1 Announce Type: new
Abstract: Sparse autoencoder (SAE)-based steering has been widely used to address knowledge conflicts by guiding LLMs to be more faithful to the contextual knowl...
By Wenbo Zhang, Zhongxiang Sun, Zhiguang Han, Jun Xu
GoDeep is an annotation‑free method for open‑vocabulary 3D scene understanding that uses a vision‑language model solely as a translator to generate structured, entity‑level descriptions of each image. These descriptions are projected and aggregated in a language‑only embedding space, eliminating the need for a 3D training corpus or domain‑specific encoder. The approach achieves competitive performance on ScanNet++ and a cultural heritage benchmark, accurately localizes out‑of‑vocabulary objects, and offers explainable, point‑level predictions.
By Thodoris Betsas, Anastasios Doulamis, Andreas Georgopoulos