Ge Li
Ge Li — researcher or builder tracked in the Angestrom contributor network.
Papers · 2
Cognitive-structured Multimodal Agent for Multimodal Understanding, Generation, and Editing
Recent unified multimodal models show a single architecture can jointly perform vision/language understanding and image generation/editing. However, they repeatedly feed all historical visual and textual inputs into a shared context window, limiting long-horizon multimodal dialogue due to visual token explosion and unreliable cross-turn referencing. We propose a Cognitive-structured Multimodal Agent that externalizes visual information into an Episodic Visual Memory and selectively reactivates relevant episodes during reasoning. The agent consists of a Perceptual Abstraction Engine for structu
Automatic Thematic Indexing of Large Literary Corpora: A Machine Learning Approach to Voltaire's Complete Works
Thematic indexing -- the practice of assigning structured conceptual labels to sections of text -- is essential to scholarly access in large-scale literary and historical editions, yet it remains a largely manual, labour-intensive process. This paper explores the application of machine learning to automatic thematic indexing, using two substantial sub-corpora of the Complete Works of Voltaire as a test case: the Essai sur les mœurs et l'esprit des nations and the Questions sur l'Encyclopédie. The task is framed as a multi-label classification problem, in which a model must assign the set of in
News · 5
Grok Build goes open source — and drops usage limits - Gagadget.com
<a href="https://news.google.com/rss/articles/CBMiigFBVV95cUxObzkxWHFZWHVhRWpaemROM2lSUTQ1OXA1QmlNUXNZUkNDSGxIeHFVemY3TTlUMThKRVd2UXBOUEh4Q3pXVnpzUlhnZ1lnQkZDekphbUZEVlduYXduVVRLcWx3eXBvYl9HdzAzcTNBdk40WlN1Sm1lZWFzc3pkcjJ0M1BXcmliWDRBeUHSAY8BQVVfeXFMUDhWb0x5a3V2WFpibU9HdzRHY1ktNW44aWxPSkM3MnQxZUtDQjhQRlVORDg4bjJxWHlpd2ljMzlPUkh1WU1Dd1RHSjA0dm13QmlEck11dlNfNWotamY5RWEyZXdyWEU3LXIwaHdCR21MQjhPelFUTzJoZm91M3p3WHZIcHBTd3ljZUhNcUV5TUk?oc=5" target="_blank">Grok Build goes open source — and drops usag
AI’s Next Competitive Advantage Lives at the Edge - The AI Journal
<a href="https://news.google.com/rss/articles/CBMieEFVX3lxTE5mQjRqMm1meHZ2R2JIWmpBYkh4Q0pMT1o2ZVh5N3lWR2paeGhlYTJ3cUNBVUtrVDdON01CdW9OUTZzQ3BDRk9xOFYzcnAzNmU2b0F4MUtyeFp4VThNRGhkN0dOTTdUNnlDNXBmcEROa0ZCa0d3SDhvTg?oc=5" target="_blank">AI’s Next Competitive Advantage Lives at the Edge</a> <font color="#6f6f6f">The AI Journal</font>
Grok Build open-sources code and resets usage limits for users - Crypto Briefing
<a href="https://news.google.com/rss/articles/CBMicEFVX3lxTE5Pd3JVNW5PS3dRdkJ1cGo1OVRDV1ZNMXFWRk1DeUJ4MmZhbU82b3FBZmdhMlpTNS13STFzTUFyN1FFU05wcmhFZVBoOWVnMDdYU0ZjbDdtV3R5ZEd4aE9jRWdIdjUwejRhdUJWT3BCUzk?oc=5" target="_blank">Grok Build open-sources code and resets usage limits for users</a> <font color="#6f6f6f">Crypto Briefing</font>
CollegeConnekt Passport: AI that turns a conversation into a full Common App profile (activities, essays, college list)
<!-- SC_OFF --><div class="md"><p>Thought this AI application might be interesting to share here given the use case.</p> <p>CollegeConnekt Passport (<a href="https://collegeconnekt.com/">https://collegeconnekt.com/</a>) is a conversational AI tool for high school students applying to college. Instead of filling out forms, you just talk — and it extracts your activities list, honors, essay angles, and college list from the conversation.</p> <p>A few things that stood out:</p> <p>• Activities are
Clifford Chance builds 400,000-document AI knowledge library with Epiq and Microsoft - The Global Legal Post
<a href="https://news.google.com/rss/articles/CBMiygFBVV95cUxPZURXX3hVRndqRndhakpzQnAxTjBQSExxMk9YUW1wb180RmdHbHFhd0cwMXZjc2Fwbk52RmVyWF9LQ2I0c1M5NFNjR1J4VzkzNzVYa29EVmt3MnZ1ZS1sZVhTUnlMbGxRbnlZR1BWbnZydzFVZ0tiMENTVHhjSGgxa2RlMEgyd1MyZGU2NlUtbTdOaU1STXVDU1E4WDUxNU9idXB3YlNOOU82X2dVRTdoS2ZNMmQ4aDdadFp1OVFLT3dvTkVSal9oZVF3?oc=5" target="_blank">Clifford Chance builds 400,000-document AI knowledge library with Epiq and Microsoft</a> <font color="#6f6f6f">The Global Legal Post</font>