📛 Evidence of Meaning in Language Models Trained on Programs
🧠 This paper shows that language models can learn meaning from program corpuses and offers a framework to study meaning acquisition and representation in them.
🐦 11
❤️ 1.5K
🔗 arxiv.org/pdf/2305.11169.pdf (https://arxiv.org/pdf/2305.11169.pdf)
https://nitter.moomoo.me/ArXivGPT/status/1660540772502863872#m
📛 Sparks of Artificial General Intelligence: Early experiments with GPT-4
🧠 GPT-4 shows near-human performance in multiple domains, possibly indicating early artificial general intelligence.
🐦 1K
❤️ 19.6K
✍️ @erichorvitz (https://nitter.moomoo.me/erichorvitz)
🔗 arxiv.org/pdf/2303.12712.pdf (https://arxiv.org/pdf/2303.12712.pdf)
https://nitter.moomoo.me/ArXivGPT/status/1660540770950995968#m
📛 Any-to-Any Generation via Composable Diffusion
🧠 CoDi is a versatile AI model that generates different output types like language, image, video, or audio from various inputs, maintaining high generation quality.
🐦 7
❤️ 177
🔗 arxiv.org/pdf/2305.11846.pdf (https://arxiv.org/pdf/2305.11846.pdf)
https://nitter.moomoo.me/ArXivGPT/status/1660540769323692032#m
📛 Neural Network Architecture Beyond Width and Depth
🧠 NestNet is a 3D neural network with enhanced expressiveness and accuracy for approximating Lipschitz continuous functions compared to 2D networks.
🐦 8
❤️ 6
🔗 arxiv.org/pdf/2205.09459.pdf (https://arxiv.org/pdf/2205.09459.pdf)
https://nitter.moomoo.me/ArXivGPT/status/1660540771689193473#m
📛 Drag Your GAN: Interactive Point-based Manipulation on the Generative
Image Manifold
🧠 DragGAN enables precise image manipulation and realistic outcomes in generative adversarial networks.
🐦 63
❤️ 6.2K
✍️ @XingangP (https://nitter.moomoo.me/XingangP)
🔗 arxiv.org/pdf/2305.10973.pdf (https://arxiv.org/pdf/2305.10973.pdf)
https://nitter.moomoo.me/ArXivGPT/status/1660540770166644736#m
📛 LIMA: Less Is More for Alignment
🧠 LIMA, a 65B parameter model, excels in unseen tasks with 1,000 prompts, suggesting pretraining imparts significant knowledge and minimal tuning is needed for quality results.
🐦 7
❤️ 467
🔗 arxiv.org/pdf/2305.11206.pdf (https://arxiv.org/pdf/2305.11206.pdf)
https://nitter.moomoo.me/ArXivGPT/status/1660540768501608450#m
📛 Towards Expert-Level Medical Question Answering with Large Language
Models
🧠 Med-PaLM 2 AI achieves near-physician accuracy (86.5%) in medical question-answering.
🐦 47
❤️ 4K
✍️ @ymatias (https://nitter.moomoo.me/ymatias)@thekaransinghal (https://nitter.moomoo.me/thekaransinghal)@vivnat (https://nitter.moomoo.me/vivnat)@alan\_karthi (https://nitter.moomoo.me/alan_karthi)
🔗 arxiv.org/pdf/2305.09617.pdf (https://arxiv.org/pdf/2305.09617.pdf)
https://nitter.moomoo.me/ArXivGPT/status/1660540767687921665#m
📛 DarkBERT: A Language Model for the Dark Side of the Internet
🧠 DarkBERT, pretrained on Dark Web data, surpasses existing models and aids in analyzing the domain's linguistic features.
🐦 39
❤️ 511
✍️ @EugeneOnNLP (https://nitter.moomoo.me/EugeneOnNLP)
🔗 arxiv.org/pdf/2305.08596.pdf (https://arxiv.org/pdf/2305.08596.pdf)
https://nitter.moomoo.me/ArXivGPT/status/1660540766068916224#m
📛 Tree of Thoughts: Deliberate Problem Solving with Large Language Models
🧠 The Tree of Thoughts framework enhances language models' problem-solving by exploring coherent text units and multiple reasoning paths.
🐦 34
❤️ 2.5K
✍️ @ShunyuYao12 (https://nitter.moomoo.me/ShunyuYao12)
🔗 arxiv.org/pdf/2305.10601.pdf (https://arxiv.org/pdf/2305.10601.pdf)
https://nitter.moomoo.me/ArXivGPT/status/1660540766882594818#m
"The generalized Hierarchical Gaussian Filter"
Cat: cs NE
Link: arxiv.org/pdf/2305.10937v1 (https://arxiv.org/pdf/2305.10937v1)
https://nitter.moomoo.me/ArXivGPT/status/1659761416864399369#m
"Improving Recommendation System Serendipity Through Lexicase Selection"
Link: arxiv.org/pdf/2305.11044v1 (https://arxiv.org/pdf/2305.11044v1)
https://nitter.moomoo.me/ArXivGPT/status/1659760906581180416#m
"Neuromorphic Bayesian Optimization in Lava"
Cat: cs NE
Link: arxiv.org/pdf/2305.11060v1 (https://arxiv.org/pdf/2305.11060v1)
https://nitter.moomoo.me/ArXivGPT/status/1659760395987595264#m
"Learning Restoration is Not Enough: Transfering Identical Mapping for Single-Image Shadow Removal"
Cat: cs CV
Link: arxiv.org/pdf/2305.10640v1 (https://arxiv.org/pdf/2305.10640v1)
https://nitter.moomoo.me/ArXivGPT/status/1659759886337736704#m
"PTQD: Accurate Post-Training Quantization for Diffusion Models"
Cat: cs CV
Link: arxiv.org/pdf/2305.10657v1 (https://arxiv.org/pdf/2305.10657v1)
https://nitter.moomoo.me/ArXivGPT/status/1659759375278559232#m
"Learning Differentially Private Probabilistic Models for Privacy-Preserving Image Generation"
Cat: cs CV
Link: arxiv.org/pdf/2305.10662v1 (https://arxiv.org/pdf/2305.10662v1)
https://nitter.moomoo.me/ArXivGPT/status/1659758864080965632#m
"Tuned Contrastive Learning"
Cat: cs CV
Link: arxiv.org/pdf/2305.10675v1 (https://arxiv.org/pdf/2305.10675v1)
https://nitter.moomoo.me/ArXivGPT/status/1659758353575469056#m
"Re-thinking Data Availablity Attacks Against Deep Neural Networks"
Link: arxiv.org/pdf/2305.10691v1 (https://arxiv.org/pdf/2305.10691v1)
https://nitter.moomoo.me/ArXivGPT/status/1659757843510349825#m
"Zero-Day Backdoor Attack against Text-to-Image Diffusion Models via Personalization"
Cat: cs CV
Link: arxiv.org/pdf/2305.10701v1 (https://arxiv.org/pdf/2305.10701v1)
https://nitter.moomoo.me/ArXivGPT/status/1659757333550100480#m
"Exploiting Fine-Grained DCT Representations for Hiding Image-Level Messages within JPEG Images"
Cat: cs CV
Link: arxiv.org/pdf/2305.06582v1 (https://arxiv.org/pdf/2305.06582v1)
https://nitter.moomoo.me/ArXivGPT/status/1657220530112716802#m
"Hyperbolic Deep Learning in Computer Vision: A Survey"
Cat: cs CV
Link: arxiv.org/pdf/2305.06611v1 (https://arxiv.org/pdf/2305.06611v1)
https://nitter.moomoo.me/ArXivGPT/status/1657219514663325698#m