BenChaliah / Superposition-Transformer
a novel architecture that leverages Autoencoders to superimpose the hidden representations of a base model and a fine-tuned model within a shared parameter space. Using B-spline-based blending coefficients and autoencoders that adaptively reconstruct the original hidden states based on the input data distribution.
☆42Updated 2 months ago
Alternatives and similar repositories for Superposition-Transformer:
Users that are interested in Superposition-Transformer are comparing it to the libraries listed below
- The official code for MedAgent_Pro☆13Updated this week
- [ICLR 2025] Weighted-Reward Preference Optimization for Implicit Model Fusion☆12Updated 2 weeks ago
- Control LLM☆13Updated this week
- [ICML 2024] Official Repository for the paper "Transformers Get Stable: An End-to-End Signal Propagation Theory for Language Models"☆9Updated 8 months ago
- This repo contains code for the paper "Both Text and Images Leaked! A Systematic Analysis of Data Contamination in Multimodal LLM"☆13Updated this week
- Official implementation of ECCV24 paper: POA☆24Updated 7 months ago
- Code for "R2-T2: Re-Routing in Test-Time for Multimodal Mixture-of-Experts"☆14Updated 3 weeks ago
- CS194-196 Course Project☆14Updated last month
- Remasking Discrete Diffusion Models with Inference-Time Scaling☆16Updated 3 weeks ago
- ☆14Updated 6 months ago
- ☆12Updated 2 months ago
- LongWriter-V: Enabling Ultra-Long and High-Fidelity Generation in Vision-Language Models☆17Updated this week
- The official repo of continuous speculative decoding☆24Updated this week
- Generative Modeling with Bayesian Sample Inference☆19Updated last week
- \infty-Video: A Training-Free Approach to Long Video Understanding via Continuous-Time Memory Consolidation☆12Updated last month
- The official implementation of ICLR 2025 paper "Polynomial Composition Activations: Unleashing the Dynamics of Large Language Models".☆13Updated last week
- NeuMeta transforms neural networks by allowing a single model to adapt on the fly to different sizes, generating the right weights when n…☆42Updated 4 months ago
- ☆11Updated 2 months ago
- Code, results and other artifacts from the paper introducing the WildChat-50m dataset and the Re-Wild model family.☆29Updated 2 months ago
- This repository contains the code for AdaCLIP, a computation and latency-aware system for pragmatic multimodal video retrieval.☆10Updated 10 months ago
- implementation of dualformer☆13Updated last month
- An End-to-End Model with Adaptive Filtering for Retrieval-Augmented Generation☆10Updated 5 months ago
- ☆13Updated 2 months ago
- A Python library for processing and filtering TabLib☆11Updated 7 months ago
- Official implementation of the paper "MMInA: Benchmarking Multihop Multimodal Internet Agents"☆41Updated last month
- [ICLR 2025] Official Pytorch Implementation of "Mix-LN: Unleashing the Power of Deeper Layers by Combining Pre-LN and Post-LN" by Pengxia…☆19Updated 3 months ago
- ☆31Updated 2 months ago
- The official implementation of Preference Data Reward-Augmentation.☆17Updated 5 months ago
- ☆20Updated 3 months ago
- Official PyTorch Implementation for Task Vectors are Cross-Modal☆22Updated 3 months ago