☆ Yσɠƚԋσʂ ☆@lemmy.ml • edit-23 months agoBy using the same techniques Google used to solve Go (MTCS and backprop), Llama8B gets 96.7% on math benchmark GSM8K. That’s better than GPT-4, Claude and Gemini, with 200x fewer parameters!plus-squareexternal-linkmessage-square0 fedilinkarrow-up17
arrow-up17external-linkBy using the same techniques Google used to solve Go (MTCS and backprop), Llama8B gets 96.7% on math benchmark GSM8K. That’s better than GPT-4, Claude and Gemini, with 200x fewer parameters!plus-square☆ Yσɠƚԋσʂ ☆@lemmy.ml • edit-23 months agomessage-square0 Commentsfedilink
☆ Yσɠƚԋσʂ ☆@lemmy.ml • 3 months agoMixture of Agents (MoA) leverages several open-source LLM agents to achieve a score of 65.1% on AlpacaEval 2.0plus-squareexternal-linkmessage-square0 fedilinkarrow-up14
arrow-up14external-linkMixture of Agents (MoA) leverages several open-source LLM agents to achieve a score of 65.1% on AlpacaEval 2.0plus-square☆ Yσɠƚԋσʂ ☆@lemmy.ml • 3 months agomessage-square0 Commentsfedilink
ylai@lemmy.ml • 3 months agoFrom DeepSpeed to FSDP and Back Again with Hugging Face Accelerateplus-squareexternal-linkmessage-square0 fedilinkarrow-up11
arrow-up11external-linkFrom DeepSpeed to FSDP and Back Again with Hugging Face Accelerateplus-squareylai@lemmy.ml • 3 months agomessage-square0 Commentsfedilink
☆ Yσɠƚԋσʂ ☆@lemmy.ml • 4 months agoSakuga-42M Dataset: Scaling Up Cartoon Researchexternal-linkmessage-square0 fedilinkarrow-up13
arrow-up13external-linkSakuga-42M Dataset: Scaling Up Cartoon Research☆ Yσɠƚԋσʂ ☆@lemmy.ml • 4 months agomessage-square0 Commentsfedilink
☆ Yσɠƚԋσʂ ☆@lemmy.ml • 4 months agoHow AI 'Understands' Images (CLIP)plus-squareexternal-linkmessage-square0 fedilinkarrow-up14
arrow-up14external-linkHow AI 'Understands' Images (CLIP)plus-square☆ Yσɠƚԋσʂ ☆@lemmy.ml • 4 months agomessage-square0 Commentsfedilink
☆ Yσɠƚԋσʂ ☆@lemmy.ml • 5 months agoHiDiffusion: Unlocking Higher-Resolution Creativity and Efficiency in Pretrained Diffusion Modelsplus-squareexternal-linkmessage-square0 fedilinkarrow-up12
arrow-up12external-linkHiDiffusion: Unlocking Higher-Resolution Creativity and Efficiency in Pretrained Diffusion Modelsplus-square☆ Yσɠƚԋσʂ ☆@lemmy.ml • 5 months agomessage-square0 Commentsfedilink
☆ Yσɠƚԋσʂ ☆@lemmy.ml • 5 months agoDynamic Typography: Bringing Text to Life via Video Diffusion Priorplus-squareexternal-linkmessage-square0 fedilinkarrow-up13
arrow-up13external-linkDynamic Typography: Bringing Text to Life via Video Diffusion Priorplus-square☆ Yσɠƚԋσʂ ☆@lemmy.ml • 5 months agomessage-square0 Commentsfedilink
☆ Yσɠƚԋσʂ ☆@lemmy.ml • 5 months agoNo "Zero-Shot" Without Exponential Data: Pretraining Concept Frequency Determines Multimodal Model Performanceexternal-linkmessage-square0 fedilinkarrow-up14
arrow-up14external-linkNo "Zero-Shot" Without Exponential Data: Pretraining Concept Frequency Determines Multimodal Model Performance☆ Yσɠƚԋσʂ ☆@lemmy.ml • 5 months agomessage-square0 Commentsfedilink
The Hobbyist@lemmy.zip • 6 months agoLooking for a specific OpenAI employee personal blogplus-squaremessage-squaremessage-square0 fedilinkarrow-up18
arrow-up18message-squareLooking for a specific OpenAI employee personal blogplus-squareThe Hobbyist@lemmy.zip • 6 months agomessage-square0 Commentsfedilink
☆ Yσɠƚԋσʂ ☆@lemmy.ml • 6 months agoIntroducing SIMA, a Scalable Instructable Multiworld Agentplus-squareexternal-linkmessage-square0 fedilinkarrow-up15
arrow-up15external-linkIntroducing SIMA, a Scalable Instructable Multiworld Agentplus-square☆ Yσɠƚԋσʂ ☆@lemmy.ml • 6 months agomessage-square0 Commentsfedilink
☆ Yσɠƚԋσʂ ☆@lemmy.ml • 6 months agoLLMs are not superintelligent | Yann LeCun and Lex Fridmanplus-squareexternal-linkmessage-square0 fedilinkarrow-up12
arrow-up12external-linkLLMs are not superintelligent | Yann LeCun and Lex Fridmanplus-square☆ Yσɠƚԋσʂ ☆@lemmy.ml • 6 months agomessage-square0 Commentsfedilink
☆ Yσɠƚԋσʂ ☆@lemmy.ml • 7 months agoSora is an AI model that can create realistic and imaginative scenes from text instructions.plus-squareexternal-linkmessage-square2 fedilinkarrow-up13
arrow-up13external-linkSora is an AI model that can create realistic and imaginative scenes from text instructions.plus-square☆ Yσɠƚԋσʂ ☆@lemmy.ml • 7 months agomessage-square2 Commentsfedilink
☆ Yσɠƚԋσʂ ☆@lemmy.ml • 7 months agoMastering Text-to-Image Diffusion: Recaptioning, Planning, and Generating with Multimodal LLMsplus-squareexternal-linkmessage-square0 fedilinkarrow-up11
arrow-up11external-linkMastering Text-to-Image Diffusion: Recaptioning, Planning, and Generating with Multimodal LLMsplus-square☆ Yσɠƚԋσʂ ☆@lemmy.ml • 7 months agomessage-square0 Commentsfedilink
☆ Yσɠƚԋσʂ ☆@lemmy.ml • 7 months agoDiffusionGPT: LLM-Driven Text-to-Image Generation Systemplus-squareexternal-linkmessage-square0 fedilinkarrow-up12
arrow-up12external-linkDiffusionGPT: LLM-Driven Text-to-Image Generation Systemplus-square☆ Yσɠƚԋσʂ ☆@lemmy.ml • 7 months agomessage-square0 Commentsfedilink
☆ Yσɠƚԋσʂ ☆@lemmy.ml • 7 months agoStable cascade uses smaller the latent space than stable diffusion resulting in faster inference and cheaper training.plus-squareexternal-linkmessage-square0 fedilinkarrow-up13
arrow-up13external-linkStable cascade uses smaller the latent space than stable diffusion resulting in faster inference and cheaper training.plus-square☆ Yσɠƚԋσʂ ☆@lemmy.ml • 7 months agomessage-square0 Commentsfedilink
☆ Yσɠƚԋσʂ ☆@lemmy.ml • 7 months agoInterpreting Neural Networks through the Polytope Lensexternal-linkmessage-square0 fedilinkarrow-up13
arrow-up13external-linkInterpreting Neural Networks through the Polytope Lens☆ Yσɠƚԋσʂ ☆@lemmy.ml • 7 months agomessage-square0 Commentsfedilink
☆ Yσɠƚԋσʂ ☆@lemmy.ml • 7 months agoMatryoshka Representation Learningexternal-linkmessage-square0 fedilinkarrow-up12
arrow-up12external-linkMatryoshka Representation Learning☆ Yσɠƚԋσʂ ☆@lemmy.ml • 7 months agomessage-square0 Commentsfedilink
ericjmorey@programming.dev • 8 months agoNumPy 2 is coming: preventing breakage, updating your codeplus-squareexternal-linkmessage-square0 fedilinkarrow-up12
arrow-up12external-linkNumPy 2 is coming: preventing breakage, updating your codeplus-squareericjmorey@programming.dev • 8 months agomessage-square0 Commentsfedilink
☆ Yσɠƚԋσʂ ☆@lemmy.ml • 10 months agoMusic can be reconstructed from human auditory cortex activity using nonlinear decoding modelsplus-squareexternal-linkmessage-square0 fedilinkarrow-up11
arrow-up11external-linkMusic can be reconstructed from human auditory cortex activity using nonlinear decoding modelsplus-square☆ Yσɠƚԋσʂ ☆@lemmy.ml • 10 months agomessage-square0 Commentsfedilink
☆ Yσɠƚԋσʂ ☆@lemmy.ml • 11 months agoHuman-like systematic generalization through a meta-learning neural networkplus-squareexternal-linkmessage-square0 fedilinkarrow-up11
arrow-up11external-linkHuman-like systematic generalization through a meta-learning neural networkplus-square☆ Yσɠƚԋσʂ ☆@lemmy.ml • 11 months agomessage-square0 Commentsfedilink