{"paper":{"title":"Using RL to Identify Divisive Perspectives Improves LLMs Abilities to Identify Communities on Social Media","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Dan Goldwasser, Nikhil Mehta","submitted_at":"2024-06-03T03:45:31Z","abstract_excerpt":"The large scale usage of social media, combined with its significant impact, has made it increasingly important to understand it. In particular, identifying user communities, can be helpful for many downstream tasks. However, particularly when models are trained on past data and tested on future, doing this is difficult.\n  In this paper, we hypothesize to take advantage of Large Language Models (LLMs), to better identify user communities. Due to the fact that many LLMs, such as ChatGPT, are fixed and must be treated as black-boxes, we propose an approach to better prompt them, by training a sm"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.00969","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2406.00969/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}