{"ID":3084857,"CreatedAt":"2026-06-05T06:46:15.197025399Z","UpdatedAt":"2026-06-07T05:00:38.846751169Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2606.05713","arxiv_id":"2606.05713","title":"Beyond Generative Decoding: Discriminative Hidden-State Readout from a Native Omni-Modal LLM for Multimodal Sentiment Analysis","abstract":"Multimodal sentiment analysis (MSA) infers human affect from language, acoustic, and visual signals. Recent methods increasingly adapt large multimodal models (LMMs) via generative readout: prompting the model to emit a sentiment score as a text string. While convenient, this ties continuous regression to discrete autoregressive decoding, incurring unmeasured costs. We revisit this readout mechanism and propose a discriminative formulation built on the Thinker module of a native omni-modal LLM (Qwen2.5-Omni-7B). Instead of text decoding, we map the final-layer hidden state of the last non-padding token to a continuous score via a lightweight regression head in a single forward pass. Using 4-bit quantization and low-rank adaptation (QLoRA), the entire 7B pipeline -- including video and audio processing -- trains on a single consumer GPU (RTX 5090, 32 GB) with 10-21 GB peak memory and 1.14% trainable parameters. Through a controlled comparison fixing the backbone, data, and LoRA configuration, we isolate the impact of the readout. On CMU-MOSI and CMU-MOSEI, our discriminative readout reaches state-of-the-art accuracy without task-specific feature engineering (MOSI: MAE 0.551, Corr 0.888; MOSEI: MAE 0.506, Corr 0.790) and exhibits strong multi-seed stability. In contrast, the generative readout -- even after equivalent supervised training -- more than doubles the mean absolute error, yields unparsable or out-of-range outputs (2.8% zero-shot), and suffers from higher latency. Modality ablations reveal a text-dominant regime on CMU-MOSI. Our findings indicate that how an LMM is read out is as consequential as how it is trained, demonstrating that a discriminative readout offers a more accurate, efficient, and reliable alternative for continuous MSA.","short_abstract":"Multimodal sentiment analysis (MSA) infers human affect from language, acoustic, and visual signals. Recent methods increasingly adapt large multimodal models (LMMs) via generative readout: prompting the model to emit a sentiment score as a text string. While convenient, this ties continuous regression to discrete auto...","url_abs":"https://arxiv.org/abs/2606.05713","url_pdf":"https://arxiv.org/pdf/2606.05713v1","authors":"[\"Bin Wen\",\"Tien-Ping Tan\"]","published":"2026-06-04T05:12:36Z","proceeding":"cs.MM","tasks":"[\"cs.MM\",\"cs.SD\",\"eess.AS\"]","methods":"[\"Large Language Model\",\"LoRA\"]","has_code":false}