The paper introduces CodePoisonRAG, a framework that poisons retrieval-augmented code generation systems by transforming benign code artifacts into malicious ones. It injects CWE-specific vulnerabilities and false safety claims into a single task-matched artifact, achieving high success rates across multiple generators and even against a defense system. The study demonstrates that attackers can target and propagate specific weaknesses without altering the underlying language model.
By Varun Gadey, Ziad Marey, Alexandra Dmitrienko
WeaveMark is a new multi‑bit watermarking scheme for large language models that improves payload capacity, extraction accuracy, and text quality by using coded payload spreading, soft‑decision error‑correcting codes, and unbiased multilayer reweighting. It also adds zero‑bit layers for reliable detection of watermark presence. Experiments demonstrate significant gains, achieving an 89.8% match rate for 32‑bit messages at 200 tokens and maintaining 86.0% accuracy under 10% substitution attacks on 16‑bit messages, far outperforming the BiMark baseline.
By Gang-Hyun Park, Ju-Hyeong Lee, Hee-Youl Kwak, Dae-Young Yun
The paper introduces a black-box membership inference attack framework tailored for fine-tuned text-to-speech models, addressing challenges in query generation and representation engineering. It evaluates five query types, finding recitation queries most effective, and uses multi-level speech embeddings with temporal alignment for fine-grained comparison. Experiments on CosyVoice2, F5-TTS, and XTTS-v2 trained on VCTK and British Dialect datasets show high privacy leakage, with speaker-level AUC above 0.80 and record-level AUC between 0.80 and 0.90.
By Kunlin Cai, Kaiyuan Zhang, Zihang Xiang, Jinghuai Zhang, Abeer Alwan, Fnu Suya, Yuan Tian
The paper introduces C$^{3}$T, a Counterfactual Causal Conversation Transformer that models sentiment shifts in social‑media conversation trees. It treats discourse moves such as denial, evidence, and toxicity as interventions, predicts node sentiment and shifts, and attributes sentiment changes to specific ancestor messages. The authors also present CaSiRe, a causal sentiment reasoning layer that enriches rumor conversation datasets with sentiment, shift, intervention, and causal‑source annotations, and demonstrate that C$^{3}$T outperforms baseline models in robustness and interpretability.
By S M Rafiuddin, Atriya Sen
CACTUS is a new backdoor attack for decentralized federated learning that uses mask‑guided, modality‑specific operators to convert label‑consistent semantic pairs into target‑directed representation shifts. By isolating trigger effects and applying them counterfactually to clean embeddings before peer aggregation, CACTUS can propagate backdoors across repeated aggregation rounds. Experiments on speech, text, tabular, and image tasks show that with 30% malicious nodes, CACTUS achieves a mean attack success rate of 51.2% on Speech Commands and the highest mean ASR among evaluated attacks on three of four modalities, with success varying by network topology and malicious‑node ratio.
By Chao Feng, Burkhard Stiller
The paper presents a systematic framework for large language model (LLM) watermarking as a provenance tool in big data ecosystems. It categorizes existing watermarking methods along four deployment dimensions—insertion point, verification authority, operational state, and transformation threat model—and aligns them with the big data principles of Volume, Velocity, Variety, Veracity, and Value. The authors introduce a readiness framework that maps four key workloads—online generation, streaming detection, transformation pipelines, and ecosystem governance—to system-level requirements such as throughput, false-positive control, robustness, cross-domain reliability, governance, and downstream utility, while highlighting gaps between benchmark performance and real-world deployment readiness.
By Huy Phan, Kieu Dang, Ojaswi Dulal, Aiham AL Shukairi, Abby Shine, Chase Garner, Phung Lai
The paper introduces TAP-PER, a prefix‑based framework that learns compact user representations for large language model personalization. By encoding user preferences into lightweight prefix embeddings and incorporating temporal signals, TAP‑PER avoids the need for heavy per‑user adapters or prompt‑serialized histories. Experiments on six LaMP tasks show that TAP‑PER outperforms both prompt‑based and model‑based baselines while using far fewer per‑user parameters, enabling scalable personalization at large user scales.
By Heng Cao, Fan Zhang, Jian Yao, Yujie Zheng, Changlin Zhao, Lu Hao, Yuxuan Wei, Wangze Ni, Huaiyu Fu, Yuqian Sun, Xuyan Mo
DuoGesture is a co‑speech gesture generation model that separates gesture synthesis into a semantic stream and a beat stream, coordinated by a Semantic Variational Information Bottleneck that decides when semantic gestures override rhythmic motion. The semantic stream uses Motion‑Grounded Semantic Conditioning, replacing word embeddings with motion‑language representations to provide motion‑aligned semantic priors for rare gesture triggers. The beat stream is regularised by an Inertial Beat Prior, an anthropometry‑weighted arm‑chain module that reduces jitter and improves rhythmic consistency. Experiments show DuoGesture outperforms strong baselines and ablations confirm the complementary roles of semantic grounding, stochastic stream selection, and biomechanical regularisation.
By Ferdinand Paar, Lanmiao Liu, Asl{\i} \"Ozy\"urek, Serge Thill, Esam Ghaleb
The paper introduces Athena for Cloud Knowledge Base, a fully offline multi‑agent Retrieval Augmented Generation (RAG) system tailored for Vodafone Idea’s Telecom SNOC environment. It combines dense E5 Large V2 embeddings, BM25 sparse retrieval, and Knowledge Graph expansion, fusing results with Weighted CombSUM, cross‑encoder reranking, and MMR to produce a diverse evidence set. The framework further verifies each chunk with an LLM before response generation, discarding weak evidence and falling back to multi‑chunk evaluation when necessary, achieving an MRR@10 of 0.910 and an EM of 78.4% on a 4,200‑document corpus.
By Harish Saragadam, Sudhanshu Sharma, Ipsha Routray
VTOS (Vision Tools Orchestration Search) is a framework that adaptively orchestrates vision foundation tools—such as open‑vocabulary detectors, segmentation models, and post‑processing operators—by jointly searching for executable solution programs and observer programs that diagnose failures and provide feedback. The observer programs feed observations into a shared VisionThoughts knowledge base, guiding subsequent searches. In two case studies—dense object counting on LVIS‑Count and zero‑shot plant‑disease segmentation on PlantSeg‑OOD—VTOS outperforms static tool pipelines and agentic visual‑programming baselines, especially in complex scenarios like dense, occluded scenes and out‑of‑distribution segmentation.
By Jinchao Ge, Lingqiao Liu, Shuwen Zhao, Lei Wang
Loom is a generative consensus framework designed for real‑world root‑cause analysis (RCA) that combines open‑form hypotheses from modular heuristics with a lightweight large language model (LLM) synthesis step. It projects hypotheses into a continuous embedding space and uses an iterative centroid‑based reweighting algorithm to resolve conflicts, producing a single consensus that is then synthesized by one LLM call. On the OpenRCA benchmark Loom matches state‑of‑the‑art autonomous agents on some datasets while achieving significantly higher efficiency—about 26× faster and 33× faster with an 8B‑parameter synthesizer.
whyItMatters":"Loom demonstrates how embedding‑space reweighting can bridge the gap between statistical rigor and expressive LLMs, enabling efficient, trustworthy RCA in industrial settings."
By Ron Begleiter, Katya Egert Berg, Gilad Saban, Gil Shabat
The paper presents a SciBERT-based method for automatically classifying scientific papers into four telescope-related categories—science, instrumentation, mention, and not telescope—within strict 512-token limits. Despite truncation challenges, the approach achieved a macro F1 score of 0.89, topping the WASP-2025 leaderboard. The authors analyze truncation effects, compare chunking and long-context models, and offer insights into efficient scientific text curation.
By Madhusudhana Naidu
SignMatch introduces a prototype‑structured embedding space that learns to match dictionary sign videos with continuous sign language footage based solely on visual similarity of handshape and motion. By mapping isolated dictionary exemplars into this space, the method enables direct, embedding‑based sign matching and can generalise to unseen signs using only dictionary examples. Experiments on ASL‑Citizen, ChaLearn OSLWL, and BOBSL CSLR2 benchmarks show strong cross‑dataset, cross‑task, and cross‑language performance, outperforming prior approaches on American, British, and Spanish sign languages without benchmark‑specific supervision.
By Ryan Wong, Youngjoon Jang, Liliane Momeni, G\"ul Varol, Andrew Zisserman
Bernini proposes a unified framework that separates semantic planning and pixel rendering for video generation and editing. An MLLM-based planner predicts target semantics in ViT embedding space, while a DiT-based renderer synthesizes pixels conditioned on this plan, text features, and source VAE features for editing. The approach introduces Segment-Aware 3D Rotary Positional Embedding and chain-of-thought reasoning, achieving state‑of‑the‑art performance on diverse video benchmarks.
By Bernini Team, Chenchen Liu, Junyi Chen, Lei Li, Lu Chi, Mingzhen Sun, Zhuoying Li, Yi Fu, Ruoyu Guo, Yiheng Wu, Ge Bai, Zehuan Yuan
ExecRetrieval is a new benchmark for code‑embedding retrieval that contains 939 Python tasks, each with a verified correct implementation and up to four single‑edit buggy distractors generated mechanically. The dataset allows direct testing of a retriever’s ability to functionally discriminate correct code from near‑clone incorrect code, rather than relying on lexical similarity. Experiments on 23 dense embeddings and BM25 show that while the best system can retrieve the correct code within the top 10 results, it often fails to rank the correct implementation first, with rank‑1 errors dominated by paired buggy variants.
By Aaryan Kapoor, Md Abdullah Al Hafiz Khan
The paper presents a decision‑support system that enhances retrieval‑augmented generation (RAG) for customer contact centers by first identifying customer questions in real time. If a query matches a frequently asked question (FAQ), the system retrieves the answer directly from the FAQ database; otherwise it generates an answer via RAG, delivering responses to agents within two seconds. The approach reduces manual query formulation, lowers average handling times, and cuts operational costs, and it includes an automated workflow that uses LLMs to extract FAQs from historical transcripts when none are predefined.
By Garima Agrawal, Sashank Gummuluri, Cosimo Spera
The paper investigates how machine learning models can regress scale‑free processes, such as earthquakes or avalanches, focusing on predicting rare, large events that require extrapolation. It studies two self‑similar systems: a 2‑dimensional fractional Gaussian field and the Abelian sandpile model. Experiments compare existing architectures (U‑net, Riesz network) with new proposals (wavelet‑based Graph Neural Network, Fourier embedding, Fourier‑Mellin Neural Operator) to identify spectral bias and coarse‑graining challenges and suggest inductive biases to address them.
By Anaclara Alvez-Canepa, Cyril Furtlehner, Fran\c{c}ois P. Landes
The paper introduces a new image retrieval framework that merges neighbor embedding projections with rank-based manifold learning via rank aggregation. It uses UMAP to create low‑dimensional feature representations and combines ranked lists from UMAP and rank‑based re‑ranking methods using the Borda Count strategy. Experiments on public datasets with ResNet152, Swin Transformer, and DINOv2 features show that this combined approach improves retrieval performance, especially in scenarios where baseline representations have low precision.
By Vinicius Atsushi Sato Kawai, Gustavo Rosseto Leticio, Lucas Pascotti Valem, Daniel Carlos Guimar\~aes Pedronette
The paper investigates whether a model that has undergone class unlearning can still recover forgotten classes without access to original data. It introduces a white‑box audit method that generates synthetic probes in representation space, filters them by confidence, and relabels boundary‑adjacent probes as the forgotten class. The authors define a Relearning Score to quantify recovery while preserving retain performance, and demonstrate that several unlearning techniques on CIFAR‑10, CIFAR‑100, and TinyImageNet can be fully recovered in a source‑free setting, sometimes even outperforming a retrained reference.
By Zahra Dehghani, Pablo Piantanida, Mohammadhadi Shateri
The paper investigates why large language models (LLMs) fail to maintain continuous mixtures of token embeddings—used in latent-state reasoning—to preserve multiple reasoning paths. Through theory and experiments, it identifies three failure sources: transformer geometry distortion, amplification or contraction dynamics from softmax and autoregressive feedback, and the need for context-dependent corrections that scale with mixture size. Empirical results confirm the predicted transition between contraction and amplification and show pretrained models largely fall on the amplifying side.
By Ali Backour