Viewing enriched Twitter post
1/ Can Large Language Models (LLMs) truly reason? Or are they just sophisticated pattern matchers? In our latest preprint, we explore this key question through a large-scale study of both open-source like Llama, Phi, Gemma, and Mistral and leading closed models, including the recent OpenAI GPT-4o and o1-series. https://t.co/2tv8Pp9MSz Work done with @i_mirzadeh, @KeivanAlizadeh2, Hooman Shahrokhi, Samy Bengio, @OncelTuzel. #LLM #Reasoning #Mathematics #AGI #Research #Apple
{
"media": [
{
"id": "",
"type": "photo",
"url": null,
"media_url": "https://pbs.twimg.com/media/GZjJoMBaUAAe0Lw.png",
"media_url_https": null,
"display_url": null,
"expanded_url": null
}
],
"nlp": {
"sentiment": "neutral",
"processed_at": "2025-08-06T12:55:32.980630"
},
"score": 1.0,
"scored_at": "2025-08-09T13:46:07.547175",
"import_source": "network_archive_import",
"score_components": {
"author": 0.09,
"engagement": 0.19571186738838708,
"quality": 0.18,
"source": 0.12,
"nlp": 0.1,
"recency": 0.020000000000000004
},
"source_tagged_at": "2025-08-09T13:43:13.334352",
"enriched": true,
"enriched_at": "2025-08-09T13:43:13.334354",
"links_checked": true,
"checked_at": "2025-08-10T10:32:38.129057",
"original_structure": "had_media_only"
}{
"user": {
"created_at": "2021-01-06T04:03:06.000Z",
"default_profile_image": false,
"description": "Research Scientist at @Apple, prev @DeepMind, prev @GeorgiaTech",
"fast_followers_count": 0,
"favourites_count": 320,
"followers_count": 6484,
"friends_count": 170,
"has_custom_timelines": false,
"is_translator": false,
"listed_count": 89,
"location": "Seattle Area",
"media_count": 35,
"name": "Mehrdad Farajtabar",
"normal_followers_count": 6484,
"possibly_sensitive": false,
"profile_image_url_https": "https://pbs.twimg.com/profile_images/1346668915301126144/G8kiuqen_normal.jpg",
"screen_name": "MFarajtabar",
"statuses_count": 138,
"translator_type": "none",
"url": "https://t.co/Z39ZU7S9Zt",
"verified": false,
"withheld_in_countries": [],
"id_str": "1346668532210176000"
},
"id": "1844456880971858028",
"conversation_id": "1844456880971858028",
"full_text": "1/ Can Large Language Models (LLMs) truly reason? Or are they just sophisticated pattern matchers? In our latest preprint, we explore this key question through a large-scale study of both open-source like Llama, Phi, Gemma, and Mistral and leading closed models, including the recent OpenAI GPT-4o and o1-series.\nhttps://t.co/2tv8Pp9MSz\n\nWork done with @i_mirzadeh, @KeivanAlizadeh2, Hooman Shahrokhi, Samy Bengio, @OncelTuzel.\n\n#LLM #Reasoning #Mathematics #AGI #Research #Apple",
"reply_count": 199,
"retweet_count": 1306,
"favorite_count": 5595,
"hashtags": [],
"symbols": [],
"user_mentions": [],
"urls": [],
"media": [
{
"media_url": "https://pbs.twimg.com/media/GZjJoMBaUAAe0Lw.png",
"type": "photo"
}
],
"url": "https://twitter.com/MFarajtabar/status/1844456880971858028",
"created_at": "2024-10-10T19:16:03.000Z",
"#sort_index": "1844456880971858028",
"view_count": 1579080,
"quote_count": 402,
"is_quote_tweet": false,
"is_retweet": false,
"is_pinned": false,
"is_truncated": true,
"startUrl": "https://x.com/mfarajtabar/status/1844456880971858028"
}