This paper introduces robustness indicators to systematically analyze how multilingual text embedding model rankings change based on dataset composition and aggregation methods, revealing that only a…

This paper investigates whether model compression techniques (like quantization and pruning) preserve a Large Language Model's ability to quantify its own uncertainty, finding that accuracy-only evalu…

This paper introduces SkMTEB, a comprehensive text embedding benchmark for Slovak, and develops efficient, locally-deployable Slovak embeddings.

The paper analyzes the robustness of current LLM watermarking schemes against various text modifications, concluding that watermarks can be removed with reasonable effort.

The paper proposes a secure and practical black-box text steganography method that uses a dynamic codebook and a multimodal LLM to embed secret messages into captions, outperforming existing technique…

Clark Hash is a stateless, deterministic quantization method that significantly reduces the storage size of neural embeddings while maintaining high accuracy for cosine similarity search.

The paper introduces Single-stage Sparse Retrieval (SSR), a method that replaces computationally expensive vector clustering with sparse autoencoding to achieve highly efficient multi-vector retrieval…

The paper proposes SubFit, a novel compression technique that achieves superior LLM compression by replacing non-contiguous, submodule-level components (Attention and FeedForward) with lightweight res…

XMark introduces a novel multi-bit watermarking technique that reliably embeds binary messages into LLM-generated text while maintaining high text quality and robust performance even with limited toke…

The paper demonstrates a class of steganographic exfiltration attacks against vector databases by hiding data within embeddings, and proposes VectorPin, a cryptographic provenance protocol to detect s…

The paper proposes an efficient and provably secure linguistic steganography method using range coding that achieves high embedding capacity and speed, outperforming existing methods.

This paper investigates the redundancy of the prompt KV cache during language model decoding, finding that the structure provided by chat templates is the primary source of redundancy, not the actual…

The paper introduces BREW, a novel framework that significantly improves the reliability of multi-bit text watermarking for LLMs by replacing flawed decoding-centric methods with a designated two-stag…

This paper compares PCA and LPC for dimensionality reduction in cyberattack classification, demonstrating that both techniques can achieve substantial feature compression with minimal loss of classifi…

This paper introduces survey sampling techniques to estimate or minimize empirical pairwise loss functions, showing that targeting informative pairs significantly reduces computational cost while main…

This paper introduces a new scaling law for sparse language models trained with limited data, demonstrating that sparsity can significantly improve performance and delay data saturation during multi-e…

The paper demonstrates that combining outputs from multiple large language models (LLMs) effectively cancels out statistical watermarks, revealing a fundamental vulnerability in current AI text detect…

The paper proposes REED, a post-training representation editing method that significantly improves cross-domain linguistic steganalysis performance by deterministically editing intermediate feature re…