6 papers
Constrained Adaptive Rejection Sampling
PaweÅ Parys, Sairam Vaidya, Taylor Berg-Kirkpatrick +1
Language Models (LMs) are increasingly used in applications where generated outputs must satisfy strict semantic or syntactic constraints. Existing approaches to constrained genera…
Learning the Error Patterns of Language Models
Jinwoo Kim, Taylor Berg-KirkPatrick, Loris D'Antoni
When generating outputs for domains with specific validity constraints (e.g., a program should compile), LLMs often fail in a small number of focused ways: for example, by using Py…
Continuous Diffusion Models Can Obey Formal Syntax
Jinwoo Kim, Taylor Berg-Kirkpatrick, Loris D'Antoni
Diffusion language models offer a promising alternative to autoregressive models due to their global, non-causal generation process, but their continuous latent dynamics make discr…
The Format Tax
Ivan Yee Lee, Loris D'Antoni, Taylor Berg-Kirkpatrick
Asking a large language model to respond in JSON should be a formatting choice, not a capability tax. Yet we find that structured output requirements -- JSON, XML, LaTeX, Markdown…
Grammar-Aligned Decoding
Kanghee Park, Jiayu Wang, Taylor Berg-Kirkpatrick +2
Large Language Models (LLMs) struggle with reliably generating highly structured outputs, such as program code, mathematical formulas, or well-formed markup. Constrained decoding a…
Constrained Sampling for Language Models Should Be Easy: An MCMC Perspective
Emmanuel Anaya Gonzalez, Sairam Vaidya, Kanghee Park +3
Constrained decoding enables Language Models (LMs) to produce samples that provably satisfy hard constraints. However, existing constrained-decoding approaches often distort the un…