Crowdsourcing the evolution of text parsing with unified

Introduction - Tree-sitter

Tree-sitter is a parser generator tool and an incremental parsing library. It can build a concrete syntax tree for a source file and efficiently update the syntax tree as the source file is edited. Tree-sitter aims to be:

Tree-sitter - a new parsing system for programming tools - Strange Loop

Strange Loop is a conference for software developers covering programming langs, databases, distributed systems, security, machine learning, creativity, and more!

Tokenization: A Survey for Modern NLP

While modern language models take raw text as their input and produce raw text as output, they do not operate over text directly. Hidden in the very first step of language model pipelines is...

Syntax Club

Share your syntax. Spread knowledge. A place to share, save, and discover code snippets on the AT Protocol.

Learning to Simplify

Over the last few years the availability of comparable or parallel corpora of original and simplified or adapted textual material has made possible a set of approaches to learn various types of simplification operations from corpora. Notably, with the Simple English Wikipedia (SEW), in combination with the “ordinary” English Wikipedia (EW), simplification approaches which use primarily machine learning techniques have been proposed. Most of them learn lexical changes in the data although some approaches also are able to learn some syntactic transformations. The chapter starts by introducing methods which cast text simplification, in different languages, as monolingual machine translation. Then, we will overview simplification as a statistical syntactic-tree translation process with the aim of transforming complex sentences into simpler ones. We will also overview works which apply optimization techniques aimed at finding the best possible simplification by either optimizing the application of rules or by overgenerating simplification solutions. Finally, to end the chapter we will review recent work on incorporating semantic information into the simplification problem.

Artificial Artificial Artificial Intelligence: Crowd Workers Widely Use Large Language Models for Text Production Tasks

Large language models (LLMs) are remarkable data annotators. They can be used to generate high-fidelity supervised training data, as well as survey and experimental data. With the widespread adoption of LLMs, human gold--standard annotations are key to understanding the capabilities of LLMs and the validity of their results. However, crowdsourcing, an important, inexpensive way to obtain human annotations, may itself be impacted by LLMs, as crowd workers have financial incentives to use LLMs to increase their productivity and income. To investigate this concern, we conducted a case study on the prevalence of LLM usage by crowd workers. We reran an abstract summarization task from the literature on Amazon Mechanical Turk and, through a combination of keystroke detection and synthetic text classification, estimate that 33-46% of crowd workers used LLMs when completing the task. Although generalization to other, less LLM-friendly tasks is unclear, our results call for platforms, researchers, and crowd workers to find new ways to ensure that human data remain human, perhaps using the methodology proposed here as a stepping stone. Code/data: https://github.com/epfl-dlab/GPTurk

The Stanford NLP Group

Performing groundbreaking Natural Language Processing research since 1999.

Back-to-basics approach can match or outperform AI in language analysis

A new study led by Dr Andrea Nini at The University of Manchester has found that a grammar-based approach to language analysis can match or outperform advanced AI systems in identifying who wrote a text. The method, called LambdaG, uses patterns in grammar and sentence construction rather than large-scale AI models, offering comparable accuracy ...

How large language models can reshape collective intelligence

Collective intelligence underpins the success of groups, organizations, markets and societies. Through distributed cognition and coordination, collectives can achieve outcomes that exceed the capabilities of individuals—even experts—resulting in improved accuracy and novel capabilities. Often, collective intelligence is supported by information technology, such as online prediction markets that elicit the ‘wisdom of crowds’, online forums that structure collective deliberation or digital platforms that crowdsource knowledge from the public. Large language models, however, are transforming how information is aggregated, accessed and transmitted online. Here we focus on the unique opportunities and challenges this transformation poses for collective intelligence. We bring together interdisciplinary perspectives from industry and academia to identify potential benefits, risks, policy-relevant considerations and open research questions, culminating in a call for a closer examination of how large language models affect humans’ ability to collectively tackle complex problems.

How large language models can reshape collective intelligence

Collective intelligence underpins the success of groups, organizations, markets and societies. Through distributed cognition and coordination, collectives can achieve outcomes that exceed the capabilities of individuals—even experts—resulting in improved accuracy and novel capabilities. Often, collective intelligence is supported by information technology, such as online prediction markets that elicit the ‘wisdom of crowds’, online forums that structure collective deliberation or digital platforms that crowdsource knowledge from the public. Large language models, however, are transforming how information is aggregated, accessed and transmitted online. Here we focus on the unique opportunities and challenges this transformation poses for collective intelligence. We bring together interdisciplinary perspectives from industry and academia to identify potential benefits, risks, policy-relevant considerations and open research questions, culminating in a call for a closer examination of how large language models affect humans’ ability to collectively tackle complex problems.

Elias Stengel-Eskin, Aaron Steven White, Sheng Zhang, Benjamin Van Durme · Universal Decompositional Semantic Parsing · SlidesLive

Professional Conference Recording

arcee-ai/mergekit

Tools for merging pretrained large language models.

An 8 Year Algorithm Nemesis: Analyzing long string trends in a body of text with suffix trees

Peritext: A CRDT for Rich-Text Collaboration

Collaboration on rich text is hard to model with plain-text approaches. We review the challenges and how to construct a CRDT for rich text.