SA-OOSC: A Multimodal LLM-Distilled Semantic Communication Framework for Enhanced Coding Efficiency with Scenario Understanding

This paper introduces SA-OOSC, a multimodal large language models (MLLM)-distilled semantic communication framework that achieves efficient semantic coding with scenario-aware importance allocations. This approach addresses a critical limitation of existing object-oriented semantic communication (OOSC) systems - assigning static importance values to specific classes of objects regardless of their contextual relevance. Our framework utilizes MLLMs to identify the scenario-augmented (SA) semantic importance for objects within the image. Through knowledge distillation with the MLLM-annotated data, our vectorization/de-vectorization networks and JSCC encoder/decoder learn to dynamically allocate coding resources based on contextual significance, i.e., distinguishing between high-importance objects and low-importance according to the SA scenario information of the task. The framework features three core innovations: a MLLM-guided knowledge distillation pipeline, an importance-weighted variable-length JSCC framework, and novel loss function designs that facilitate the knowledge distillation within the JSCC framework. Experimental validation demonstrates our framework's superior coding efficiency over conventional semantic communication systems, with open-sourced MLLM-annotated and human-verified datasets established as new benchmarks for future research in semantic communications.

ReSURE: Regularizing Supervision Unreliability for Multi-turn Dialogue Fine-tuning

Fine-tuning multi-turn dialogue systems requires high-quality supervision but often suffers from degraded performance when exposed to low-quality data. Supervision errors in early turns can propagate across subsequent turns, undermining coherence and response quality. Existing methods typically address data quality via static prefiltering, which decouples quality control from training and fails to mitigate turn-level error propagation. In this context, we propose ReSURE (Regularizing Supervision UnREliability), an adaptive learning method that dynamically down-weights unreliable supervision without explicit filtering. ReSURE estimates per-turn loss distributions using Welford's online statistics and reweights sample losses on the fly accordingly. Experiments on both single-source and mixed-quality datasets show improved stability and response quality. Notably, ReSURE enjoys positive Spearman correlations (0.21 ~ 1.0 across multiple benchmarks) between response scores and number of samples regardless of data quality, which potentially paves the way for utilizing large-scale data effectively. Code is publicly available at https://github.com/Elvin-Yiming-Du/ReSURE_Multi_Turn_Training.

ReGraP-LLaVA: Reasoning enabled Graph-based Personalized Large Language and Vision Assistant

Recent advances in personalized MLLMs enable effective capture of user-specific concepts, supporting both recognition of personalized concepts and contextual captioning. However, humans typically explore and reason over relations among objects and individuals, transcending surface-level information to achieve more personalized and contextual understanding. To this end, existing methods may face three main limitations: Their training data lacks multi-object sets in which relations among objects are learnable. Building on the limited training data, their models overlook the relations between different personalized concepts and fail to reason over them. Their experiments mainly focus on a single personalized concept, where evaluations are limited to recognition and captioning tasks. To address the limitations, we present a new dataset named ReGraP, consisting of 120 sets of personalized knowledge. Each set includes images, KGs, and CoT QA pairs derived from the KGs, enabling more structured and sophisticated reasoning pathways. We propose ReGraP-LLaVA, an MLLM trained with the corresponding KGs and CoT QA pairs, where soft and hard graph prompting methods are designed to align KGs within the model's semantic space. We establish the ReGraP Benchmark, which contains diverse task types: multiple-choice, fill-in-the-blank, True/False, and descriptive questions in both open- and closed-ended settings. The proposed benchmark is designed to evaluate the relational reasoning and knowledge-connection capability of personalized MLLMs. We conduct experiments on the proposed ReGraP-LLaVA and other competitive MLLMs. Results show that the proposed model not only learns personalized knowledge but also performs relational reasoning in responses, achieving the SoTA performance compared with the competitive methods. All the codes and datasets are released at: https://github.com/xyfyyds/ReGraP.