Nlp
Fuel prices are soaring. Plastic could be next. | MIT Technology Review
The war’s economic effects are hitting all sorts of fossil-derived products.
MIT Technology Review - AI
·
6 min
·
Natural Language Processing
Text understanding and language tasks
Top This Week
![[2602.00750] Bypassing Prompt Injection Detectors through Evasive Injections](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
Llms
[2602.00750] Bypassing Prompt Injection Detectors through Evasive Injections
Abstract page for arXiv paper 2602.00750: Bypassing Prompt Injection Detectors through Evasive Injections
arXiv - AI
·
4 min
·
Nlp
[2512.18640] Geometric-Photometric Event-based 3D Gaussian Ray Tracing
Abstract page for arXiv paper 2512.18640: Geometric-Photometric Event-based 3D Gaussian Ray Tracing
arXiv - AI
·
4 min
·
All Content
Llms
[2602.22424] Causality $\neq$ Invariance: Function and Concept Vectors in LLMs
This paper investigates the representation of concepts in large language models (LLMs), revealing that Function Vectors (FVs) are not ful...
arXiv - Machine Learning
·
4 min
·
Nlp
[2602.22282] Differentially Private Truncation of Unbounded Data via Public Second Moments
This paper presents a novel approach to differentially private data truncation using public second moments, enhancing privacy without com...
arXiv - Machine Learning
·
4 min
·
Llms
[2602.22278] RETLLM: Training and Data-Free MLLMs for Multimodal Information Retrieval
The paper presents RETLLM, a novel framework for multimodal information retrieval (MMIR) that operates without the need for training or l...
arXiv - Machine Learning
·
4 min
·
Nlp
[2602.22275] Deep Accurate Solver for the Geodesic Problem
This article presents a novel deep learning approach for accurately solving the geodesic problem on continuous surfaces, achieving third-...
arXiv - Machine Learning
·
4 min
·
Llms
[2602.22368] EyeLayer: Integrating Human Attention Patterns into LLM-Based Code Summarization
The paper presents EyeLayer, a novel module that integrates human attention patterns into LLM-based code summarization, enhancing model p...
arXiv - AI
·
4 min
·
Llms
[2602.22359] Scaling In, Not Up? Testing Thick Citation Context Analysis with GPT-5 and Fragile Prompts
This paper explores the effectiveness of GPT-5 in interpretative citation context analysis (CCA) by employing thick, text-grounded readin...
arXiv - AI
·
4 min
·
Llms
[2602.22246] Self-Purification Mitigates Backdoors in Multimodal Diffusion Language Models
This article presents a framework called DiSP (Diffusion Self-Purification) to mitigate backdoor attacks in Multimodal Diffusion Language...
arXiv - Machine Learning
·
4 min
·
Llms
[2602.22351] Decoder-based Sense Knowledge Distillation
This paper introduces Decoder-based Sense Knowledge Distillation (DSKD), a novel framework that enhances knowledge distillation in decode...
arXiv - AI
·
3 min
·
Llms
[2602.22299] Decoding the Hook: A Multimodal LLM Framework for Analyzing the Hooking Period of Video Ads
This article presents a framework using multimodal large language models (MLLMs) to analyze the 'hooking period' of video ads, focusing o...
arXiv - Machine Learning
·
4 min
·
Llms
[2602.22223] SQaLe: A Large Text-to-SQL Corpus Grounded in Real Schemas
The paper introduces SQaLe, a large-scale text-to-SQL dataset designed to enhance the development of models that convert natural language...
arXiv - Machine Learning
·
3 min
·
Llms
[2602.23353] SOTAlign: Semi-Supervised Alignment of Unimodal Vision and Language Models via Optimal Transport
The paper introduces SOTAlign, a semi-supervised framework for aligning unimodal vision and language models using minimal paired data and...
arXiv - AI
·
4 min
·
Ai Infrastructure
[2602.23320] ParamMem: Augmenting Language Agents with Parametric Reflective Memory
The paper introduces ParamMem, a parametric memory module designed to enhance language agents by enabling diverse reflective outputs, imp...
arXiv - Machine Learning
·
3 min
·
Machine Learning
[2602.23296] Conformalized Neural Networks for Federated Uncertainty Quantification under Dual Heterogeneity
This article presents FedWQ-CP, a novel approach to federated uncertainty quantification that addresses dual heterogeneity in data and mo...
arXiv - Machine Learning
·
4 min
·
Nlp
[2602.23280] Physics Informed Viscous Value Representations
This paper presents a novel approach to offline goal-conditioned reinforcement learning by introducing a physics-informed regularization ...
arXiv - Machine Learning
·
3 min
·
Llms
[2602.22240] From Prompts to Performance: Evaluating LLMs for Task-based Parallel Code Generation
This paper evaluates the performance of Large Language Models (LLMs) in generating task-based parallel code using various input prompts a...
arXiv - AI
·
3 min
·
Nlp
[2602.22237] Optimized Disaster Recovery for Distributed Storage Systems: Lightweight Metadata Architectures to Overcome Cryptographic Hashing Bottleneck
This paper presents a novel approach to disaster recovery in distributed storage systems, addressing the limitations of cryptographic has...
arXiv - AI
·
3 min
·
Machine Learning
[2602.23201] Tell Me What To Learn: Generalizing Neural Memory to be Controllable in Natural Language
This paper presents a generalized neural memory system that allows for flexible updates based on natural language instructions, addressin...
arXiv - Machine Learning
·
3 min
·
Llms
[2602.22225] SmartChunk Retrieval: Query-Aware Chunk Compression with Planning for Efficient Document RAG
The paper presents SmartChunk Retrieval, a query-aware framework that enhances retrieval-augmented generation (RAG) by adapting chunk siz...
arXiv - Machine Learning
·
4 min
·
Machine Learning
[2602.22224] DS SERVE: A Framework for Efficient and Scalable Neural Retrieval
DS SERVE is a framework designed to enhance neural retrieval systems by efficiently processing large-scale text datasets, achieving low l...
arXiv - AI
·
3 min
·
Llms
[2602.22219] Comparative Analysis of Neural Retriever-Reranker Pipelines for Retrieval-Augmented Generation over Knowledge Graphs in E-commerce Applications
This article presents a comparative analysis of neural retriever-reranker pipelines for retrieval-augmented generation (RAG) in e-commerc...
arXiv - AI
·
4 min
·
Related Topics
Stay updated with AI News
Get the latest news, tools, and insights delivered to your inbox.
Daily or weekly digest • Unsubscribe anytime