Saibo Geng
YOU?
Author Swipe
View article: TRPrompt: Bootstrapping Query-Aware Prompt Optimization from Textual Rewards
TRPrompt: Bootstrapping Query-Aware Prompt Optimization from Textual Rewards Open
Prompt optimization improves the reasoning abilities of large language models (LLMs) without requiring parameter updates to the target model. Following heuristic-based "Think step by step" approaches, the field has evolved in two main dire…
View article: zip2zip: Inference-Time Adaptive Tokenization via Online Compression
zip2zip: Inference-Time Adaptive Tokenization via Online Compression Open
Tokenization efficiency plays a critical role in the performance and cost of large language models (LLMs), yet most models rely on static tokenizers optimized on general-purpose corpora. These tokenizers' fixed vocabularies often fail to a…
View article: JSONSchemaBench: A Rigorous Benchmark of Structured Outputs for Language Models
JSONSchemaBench: A Rigorous Benchmark of Structured Outputs for Language Models Open
Reliably generating structured outputs has become a critical capability for modern language model (LM) applications. Constrained decoding has emerged as the dominant technology across sectors for enforcing structured outputs during generat…
View article: Byte BPE Tokenization as an Inverse string Homomorphism
Byte BPE Tokenization as an Inverse string Homomorphism Open
Tokenization is an important preprocessing step in the training and inference of large language models (LLMs). While there has been extensive research on the expressive power of the neural achitectures used in LLMs, the impact of tokenizat…
View article: Sketch-Guided Constrained Decoding for Boosting Blackbox Large Language Models without Logit Access
Sketch-Guided Constrained Decoding for Boosting Blackbox Large Language Models without Logit Access Open
Constrained decoding, a technique for enforcing constraints on language model outputs, offers a way to control text generation without retraining or architectural modifications. Its application is, however, typically restricted to models t…
View article: Flows: Building Blocks of Reasoning and Collaborating AI
Flows: Building Blocks of Reasoning and Collaborating AI Open
Recent advances in artificial intelligence (AI) have produced highly capable and controllable systems. This creates unprecedented opportunities for structured reasoning as well as collaboration among multiple AI systems and humans. To full…
View article: Grammar-Constrained Decoding for Structured NLP Tasks without Finetuning
Grammar-Constrained Decoding for Structured NLP Tasks without Finetuning Open
Despite their impressive performance, large language models (LMs) still struggle with reliably generating complex output structures when not finetuned to follow the required output format exactly. To address this issue, grammar-constrained…
View article: Grammar-Constrained Decoding for Structured NLP Tasks without Finetuning
Grammar-Constrained Decoding for Structured NLP Tasks without Finetuning Open
International audience
View article: Legal Transformer Models May Not Always Help
Legal Transformer Models May Not Always Help Open
Deep learning-based Natural Language Processing methods, especially transformers, have achieved impressive performance in the last few years. Applying those state-of-the-art NLP methods to legal activities to automate or simplify some simp…
View article: An Enhanced MeanSum Method For Generating Hotel Multi-Review Summarizations
An Enhanced MeanSum Method For Generating Hotel Multi-Review Summarizations Open
Multi-document summaritazion is the process of taking multiple texts as input and producing a short summary text based on the content of input texts. Up until recently, multi-document summarizers are mostly supervised extractive. However, …