Papers by Jintao Cheng

4 papers
Training LLMs to be Better Text Embedders through Bidirectional Reconstruction (2025.emnlp-main)

Copied to clipboard

Challenge: Existing text embedding approaches often leverage the embeddment of the final token, typically a reserved special token such as ‘[EOS]‘.
Approach: They propose to add a new training stage before contrastive learning to enrich the semantics of the final token embedding.
Outcome: The proposed training stage improves performance on the Massive Text Embedding Benchmark (MTEB), achieving new state-of-the-art results across different LLM base models and scales.
Uni-Dubbing: Zero-Shot Speech Synthesis from Visual Articulation (2024.acl-long)

Copied to clipboard

Challenge: Multimodal speech synthesis is a key challenge due to the scarcity of datasets that pair audio with corresponding video.
Approach: They propose a method that incorporates modality alignment during the pre-training phase on multimodal datasets and freezes the video modality extraction component and the encoder module within the pretrained weights.
Outcome: The proposed method achieves a reduced word error rate (WER) of 31.73%, surpassing the previous best of 33.9% with single-modality audio.
TSPO: Breaking the Double Homogenization Dilemma in Multi-turn Search Policy Optimization (2026.findings-acl)

Copied to clipboard

Challenge: Large Language Models (LLMs) can solve complex tasks through iterative information retrieval.
Approach: They propose a turn-level stage-aware policy optimization approach to solve this problem . they introduce a first-occurrence latent reward mechanism to allocate partial rewards .
Outcome: Experiments show that TSPO outperforms state-of-the-art models on Qwen2.5-3B and 7B models.
Anchoring the Affective Manifold: Learning Canonical and Disentangled Representations via Generative Cross-Modal Alignment (2026.acl-long)

Copied to clipboard

Challenge: Dominant multimodal emotion recognition paradigms neglect the intrinsic geometric structure of affect, resulting in representations heavily entangled with non-affective factors.
Approach: They propose a Canonical Disentangled Multimodal Generative Framework that decomposes the latent space into a canonical Shared Affective Subspace and a private Modality Subspace.
Outcome: The proposed model disentangles affect from private attributes while enabling controllable emotion generation.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations