H.265/HEVC Video Steganalysis Based on CU Block Structure Gradients and IPM Mapping

Existing H.265/HEVC video steganalysis research mainly focuses on statistical feature modeling at the levels of motion vectors (MV), intra prediction modes (IPM), or transform coefficients. In contrast, studies targeting the coding-structure level - especially the analysis of block-level steganographic behaviors in Coding Units (CUs) - remain at an early stage. As a core component of H.265/HEVC coding decisions, the CU partition structure often exhibits steganographic perturbations in the form of structural changes and reorganization of prediction relationships, which are difficult to characterize effectively using traditional pixel-domain features or mode statistics. To address this issue, this paper, for the first time from the perspective of CU block-level steganalysis, proposes an H.265/HEVC video steganalysis method based on CU block-structure gradients and intra prediction mode mapping. The proposed method constructs a CU block-structure gradient map to explicitly describe changes in coding-unit partitioning, and combines it with a block-level mapping representation of IPM to jointly model the structural perturbations introduced by CU-level steganographic embedding. On this basis, we design a Transformer network, GradIPMFormer, tailored for CU-block steganalysis, thereby effectively enhancing the capability to perceive CU-level steganographic behaviors. Experimental results show that under different quantization parameters and resolution settings, the proposed method consistently achieves superior detection performance across multiple H.265/HEVC steganographic algorithms, validating the feasibility and effectiveness of conducting video steganalysis from the coding-structure perspective. This study provides a new CU block-level analysis paradigm for H.265/HEVC video steganalysis and has significant research value for covert communication security detection.

A Fine-tuning Dataset and Benchmark for Large Language Models for Protein Understanding

The parallels between protein sequences and natural language in their sequential structures have inspired the application of large language models (LLMs) to protein understanding. Despite the success of LLMs in NLP, their effectiveness in comprehending protein sequences remains an open question, largely due to the absence of datasets linking protein sequences to descriptive text. Researchers have then attempted to adapt LLMs for protein understanding by integrating a protein sequence encoder with a pre-trained LLM. However, this adaptation raises a fundamental question: "Can LLMs, originally designed for NLP, effectively comprehend protein sequences as a form of language?" Current datasets fall short in addressing this question due to the lack of a direct correlation between protein sequences and corresponding text descriptions, limiting the ability to train and evaluate LLMs for protein understanding effectively. To bridge this gap, we introduce ProteinLMDataset, a dataset specifically designed for further self-supervised pretraining and supervised fine-tuning (SFT) of LLMs to enhance their capability for protein sequence comprehension. Specifically, ProteinLMDataset includes 17.46 billion tokens for pretraining and 893,000 instructions for SFT. Additionally, we present ProteinLMBench, the first benchmark dataset consisting of 944 manually verified multiple-choice questions for assessing the protein understanding capabilities of LLMs. ProteinLMBench incorporates protein-related details and sequences in multiple languages, establishing a new standard for evaluating LLMs' abilities in protein comprehension. The large language model InternLM2-7B, pretrained and fine-tuned on the ProteinLMDataset, outperforms GPT-4 on ProteinLMBench, achieving the highest accuracy score. The dataset and the benchmark are available at https://huggingface.co/datasets/tsynbio/ProteinLMBench.