about
Stories from October 3, 2023 (UTC)
Go back a day, month, or year. Go forward a day.
1. StreamingLLM: Efficient streaming technique enable infinite sequence lengths (arxiv.org)
118 points by TheJCDenton on Oct 3, 2023 | hide | past | pdf | 12 comments
2. Merge, Then Compress: Demystify Efficient SMoE with Hints from Routing Policy (arxiv.org)
2 points by Jimmc414 on Oct 3, 2023 | hide | past | pdf | 1 comment
3. NeRF-Enhanced Outpainting for Faithful Field-of-View Extrapolation (arxiv.org)
2 points by PaulHoule on Oct 3, 2023 | hide | past | pdf | discuss
4. Understanding LLMs Through the Problem They Are Trained to Solve (arxiv.org)
2 points by JackRumford on Oct 3, 2023 | hide | past | pdf | 2 comments
5. Multi-Modal Face Stylization with a Generative Prior (arxiv.org)
1 point by PaulHoule on Oct 3, 2023 | hide | past | pdf | discuss
6. Preliminary Explorations with GPT-4V(ision) (arxiv.org)
1 point by kennethologist on Oct 3, 2023 | hide | past | pdf | discuss
7. Towards Robust Continual Learning with Bayesian Adaptive Moment Regularization (arxiv.org)
1 point by JackRumford on Oct 3, 2023 | hide | past | pdf | discuss
8. GPT-Driver: Learning to Drive with GPT (arxiv.org)
1 point by belter on Oct 3, 2023 | hide | past | pdf | 1 comment
9. PP-Met: A Real-World Personalized Prompt Based Meeting Transcription System (arxiv.org)
1 point by PaulHoule on Oct 3, 2023 | hide | past | pdf | discuss
10. The Dawn of LMMs: Preliminary Explorations with GPT-4V(ision) (arxiv.org)
1 point by vok on Oct 3, 2023 | hide | past | pdf | discuss