#semantic alignment

try —

8 papers match

cs.CV2026

ROAD: Reciprocal-Objective Alignment of Discriminative Semantics for 3D Shape Generation

Xiao Luo, Mingyang Du, Xin Zhou +5

The paper introduces ROAD, a framework that transfers semantic and structural knowledge from discriminative 3D foundation models into diffusion transformers for 3D shape generation…

#3d shape generation#diffusion models#foundation model transfer#semantic alignment
cs.CL2026

Semantic-Aligned Structural Abstraction for Multimodal Sentiment Analysis

Wei Chen, Junkai Li, Tongguan Wang +4

The paper introduces SentiLLM, a framework that converts non‑verbal signals into text‑like semantic tokens using a dual‑stream salience‑context mechanism, enabling large language m…

#multimodal sentiment analysis#large language models#semantic alignment#dual-stream calibration
cs.CV2026

Anchoring and Steering Diffusion: Enhancing the Faithfulness of Text-to-Image Generation at Inference Time

Xinyi Wang, Yuyang Huang, Yalin Su +4

The paper introduces AnchorSteer, a training‑free method that improves text‑to‑image diffusion models by initializing with CLIP‑aligned latent noise and actively correcting semanti…

#text-to-image generation#diffusion models#semantic alignment#inference-time control
cs.SD2026

From Semantics to Readout: Mechanistic Understanding of Audio Tokens after Fine-Tuning for Temporal Audio Grounding

Yujian Ma, Jinqiu Sang, Ruizhe Li +2

The paper studies how fine‑tuning large audio‑language models for temporal audio grounding alters the semantics and decoder accessibility of native audio tokens, finding that fine‑…

#audio token analysis#temporal audio grounding#model fine-tuning#decoder readability
cs.SD2026

Echoes: A semantically-aligned music deepfake detection dataset

Octavian Pascu, Dan Oneata, Horia Cucu +1

The paper presents Echoes, a new dataset of AI‑generated and real music tracks designed for robust deepfake detection, featuring semantic alignment between spoofed audio and genuin…

#music deepfake detection#audio dataset#semantic alignment#AI‑generated music
cs.CL2026

Hierarchical Acoustic-Semantic Modeling: Modality Separation and Semantic Coherence for Full-Duplex SLMs

Zhenyu Liu, Xuanyu Zhang, Yunxin Li +10

The paper identifies gradient conflicts between acoustic and semantic modeling as the cause of modality interference in full-duplex spoken language models and proposes Lychee-FD, a…

#full-duplex spoken language models#modality interference#hierarchical parameter separation#semantic alignment