Yizhe Zhang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Yizhe Zhang has a total of 34 co-authors. Here are the top co-authors by number of shared publications:
Jiatao Gu
2 shared publications- Katherine Metcalf
Apple
2 shared publications - Lucie Charlotte Magister
University of Cambridge
2 shared publications - Maartje Ter Hoeve
Apple
2 shared publications Navdeep Jaitly
2 shared publications- Navdeep Jaitly
Apple Inc
2 shared publications
Latest Publications
- Open Access
Continuously Augmented Discrete Diffusion model for Categorical Generative Modeling
Standard discrete diffusion models treat all unobserved states identically by mapping them to an absorbing [MASK] token. This creates an 'information void' where semantic information that could be inferred from unmasked tokens is lost between denoising steps. We introduce Continuously Augmented Discrete Diffusion (CADD), a framework that augments the discrete state space with a paired diffusion in a continuous latent space. This yields graded, gradually corrupted states in which masked tokens are represented by noisy yet informative latent vectors rather than collapsed 'information voids'. At each reverse step, CADD may leverage the continuous latent as a semantic hint to guide discrete denoising. The design is clean and compatible with existing discrete diffusion training. At sampling time, the strength and choice of estimator for the continuous latent vector enables a controlled trade-off between mode-coverage (generating diverse outputs) and mode-seeking (generating contextually precise outputs) behaviors. Empirically, we demonstrate CADD improves generative quality over mask-based diffusion across text generation, image synthesis, and code modeling, with consistent gains on both qualitative and quantitative metrics against strong discrete baselines.
Open AccessContinuously Augmented Discrete Diffusion model for Categorical Generative Modeling
Standard discrete diffusion models treat all unobserved states identically by mapping them to an absorbing [MASK] token. This creates an 'information void' where semantic information that could be inferred from unmasked tokens is lost between denoising steps. We introduce Continuously Augmented Discrete Diffusion (CADD), a framework that augments the discrete state space with a paired diffusion in a continuous latent space. This yields graded, gradually corrupted states in which masked tokens are represented by noisy yet informative latent vectors rather than collapsed 'information voids'. At each reverse step, CADD may leverage the continuous latent as a semantic hint to guide discrete denoising. The design is clean and compatible with existing discrete diffusion training. At sampling time, the strength and choice of estimator for the continuous latent vector enables a controlled trade-off between mode-coverage (generating diverse outputs) and mode-seeking (generating contextually precise outputs) behaviors. Empirically, we demonstrate CADD improves generative quality over mask-based diffusion across text generation, image synthesis, and code modeling, with consistent gains on both qualitative and quantitative metrics against strong discrete baselines.
- Open Access
- 1 January 2025
- Published by Association for Computational Linguistics (ACL)
- p. 3176-3192
Eliciting In-context Retrieval and Reasoning for Long-context Large Language Models
Yifu Qiu, Varun R. Embar, Yizhe Zhang, Navdeep Jaitly, Shay B Cohen, Benjamin Han. Findings of the Association for Computational Linguistics: ACL 2025. 2025.
- 1 January 2025
- Published by Association for Computational Linguistics (ACL)
- p. 3176-3192
Open AccessEliciting In-context Retrieval and Reasoning for Long-context Large Language Models
Yifu Qiu, Varun R. Embar, Yizhe Zhang, Navdeep Jaitly, Shay B Cohen, Benjamin Han. Findings of the Association for Computational Linguistics: ACL 2025. 2025.
- Open Access
- 1 January 2025
- Published by Association for Computational Linguistics (ACL)
- p. 61-77
On the Way to LLM Personalization: Learning to Remember User Conversations
Lucie Charlotte Magister, Katherine Metcalf, Yizhe Zhang, Maartje Ter Hoeve. Proceedings of the First Workshop on Large Language Model Memorization (L2M2). 2025.
- 1 January 2025
- Published by Association for Computational Linguistics (ACL)
- p. 61-77
Open AccessOn the Way to LLM Personalization: Learning to Remember User Conversations
Lucie Charlotte Magister, Katherine Metcalf, Yizhe Zhang, Maartje Ter Hoeve. Proceedings of the First Workshop on Large Language Model Memorization (L2M2). 2025.
- Top Cited PapersOpen Access
- 1 January 2025
- Published by Association for Computational Linguistics (ACL)
- p. 1160-1183
ToolSandbox: A Stateful, Conversational, Interactive Evaluation Benchmark for LLM Tool Use Capabilities
Jiarui Lu, Thomas Holleis, Yizhe Zhang, Bernhard Aumayer, Feng Nan, Haoping Bai, Shuang Ma, Shen Ma, Mengyu Li, Guoli Yin, Zirui Wang, Ruoming Pang. Findings of the Association for Computational Linguistics: NAACL 2025. 2025.
- 1 January 2025
- Published by Association for Computational Linguistics (ACL)
- p. 1160-1183
Top Cited PapersOpen AccessToolSandbox: A Stateful, Conversational, Interactive Evaluation Benchmark for LLM Tool Use Capabilities
Jiarui Lu, Thomas Holleis, Yizhe Zhang, Bernhard Aumayer, Feng Nan, Haoping Bai, Shuang Ma, Shen Ma, Mengyu Li, Guoli Yin, Zirui Wang, Ruoming Pang. Findings of the Association for Computational Linguistics: NAACL 2025. 2025.
- Open Access
On the Way to LLM Personalization: Learning to Remember User Conversations
Large Language Models (LLMs) have quickly become an invaluable assistant for a variety of tasks. However, their effectiveness is constrained by their ability to tailor responses to human preferences and behaviors via personalization. Prior work in LLM personalization has largely focused on style transfer or incorporating small factoids about the user, as knowledge injection remains an open challenge. In this paper, we explore injecting knowledge of prior conversations into LLMs to enable future work on less redundant, personalized conversations. We identify two real-world constraints: (1) conversations are sequential in time and must be treated as such during training, and (2) per-user personalization is only viable in parameter-efficient settings. To this aim, we propose PLUM, a pipeline performing data augmentation for up-sampling conversations as question-answer pairs, that are then used to finetune a low-rank adaptation adapter with a weighted cross entropy loss. Even in this first exploration of the problem, we perform competitively with baselines such as RAG, attaining an accuracy of 81.5% across 100 conversations.
Open AccessOn the Way to LLM Personalization: Learning to Remember User Conversations
Large Language Models (LLMs) have quickly become an invaluable assistant for a variety of tasks. However, their effectiveness is constrained by their ability to tailor responses to human preferences and behaviors via personalization. Prior work in LLM personalization has largely focused on style transfer or incorporating small factoids about the user, as knowledge injection remains an open challenge. In this paper, we explore injecting knowledge of prior conversations into LLMs to enable future work on less redundant, personalized conversations. We identify two real-world constraints: (1) conversations are sequential in time and must be treated as such during training, and (2) per-user personalization is only viable in parameter-efficient settings. To this aim, we propose PLUM, a pipeline performing data augmentation for up-sampling conversations as question-answer pairs, that are then used to finetune a low-rank adaptation adapter with a weighted cross entropy loss. Even in this first exploration of the problem, we perform competitively with baselines such as RAG, attaining an accuracy of 81.5% across 100 conversations.
- Open Access
- 1 January 2024
- Published by Association for Computational Linguistics (ACL)
- p. 2572-2585
Divide-or-Conquer? Which Part Should You Distill Your LLM?
Zhuofeng Wu, Richard He Bai, Aonan Zhang, Jiatao Gu, V.G.Vinod Vydiswaran, Navdeep Jaitly, Yizhe Zhang. Findings of the Association for Computational Linguistics: EMNLP 2024. 2024.
- 1 January 2024
- Published by Association for Computational Linguistics (ACL)
- p. 2572-2585
Open AccessDivide-or-Conquer? Which Part Should You Distill Your LLM?
Zhuofeng Wu, Richard He Bai, Aonan Zhang, Jiatao Gu, V.G.Vinod Vydiswaran, Navdeep Jaitly, Yizhe Zhang. Findings of the Association for Computational Linguistics: EMNLP 2024. 2024.
- Top Cited PapersOpen Access
- 1 January 2024
- Published by Association for Computational Linguistics (ACL)
- p. 14044-14072
Rephrasing the Web: A Recipe for Compute and Data-Efficient Language Modeling
Pratyush Maini, Skyler Seto, Richard Bai, David Grangier, Yizhe Zhang, Navdeep Jaitly. Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2024.
- 1 January 2024
- Published by Association for Computational Linguistics (ACL)
- p. 14044-14072
Top Cited PapersOpen AccessRephrasing the Web: A Recipe for Compute and Data-Efficient Language Modeling
Pratyush Maini, Skyler Seto, Richard Bai, David Grangier, Yizhe Zhang, Navdeep Jaitly. Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2024.