Finding Truth in LLMs: UC Berkeley & Peking U Propose Unsupervised Contrast-Consistent Search https://
syncedreview.com/2022/12/14/fin
ding-truth-in-llms-uc-berkeley-peking-u-propose-unsupervised-contrast-consistent-search/
…
@jiqizhixin
-
UC Berkeley and Peking U Propose Unsupervised Contrast-Consistent Search
By
–
-
Microsoft E5 Text Embedding Model Tops MTEB Benchmark
By
–
Microsoft’s E5 Text Embedding Model Tops the MTEB Benchmark With 40x Fewer Parameters https://
syncedreview.com/2022/12/13/mic
rosofts-e5-text-embedding-model-tops-the-mteb-benchmark-with-40x-fewer-parameters/
… -
ServiceNow and Hugging Face Release The Stack Dataset
By
–
ServiceNow Research & Hugging Face Release The Stack: 3 TB of Permissively Licensed Source Code for LLMs https://
syncedreview.com/2022/12/12/ser
vicenow-research-hugging-face-release-the-stack-3-tb-of-permissively-licensed-source-code-for-llms/
… -
Geoffrey Hinton’s Forward-Forward Algorithm Revolutionizes Neural Network Training
By
–
Geoffrey Hinton’s Forward-Forward Algorithm Charts a New Path for Neural Networks
-
Google and Lund University’s Optimus Learns Optimization Architecture
By
–
Google & Lund U’s Optimus Learned Optimization Architecture Efficiently Captures Complex Dependencies https://
syncedreview.com/2022/12/07/goo
gle-lund-us-optimus-learned-optimization-architecture-efficiently-captures-complex-dependencies/
… -
DeepMind UCL Fine-tune 70B LM for Human-aligned Statements
By
–
DeepMind & UCL Fine-tune a 70B Parameter LM to Generate Statements Agreeable to Humans with Diverse Opinions
-
Alibaba’s VQRF Achieves 100x Compression for Volumetric Files
By
–
Alibaba’s VQRF Realizes a 100x Compression Rate, Reducing Volumetric Radiance Files to 1 MB https://
syncedreview.com/2022/12/05/ali
babas-vqrf-realizes-a-100x-compression-rate-reducing-volumetric-radiance-files-to-1-mb/
… -
Stanford and Google Improve Transformer Training with Convex Analysis
By
–
Stanford U & Google’s Convex Analytic Training Framework Improves the Understanding and Optimization of Transformers https://
syncedreview.com/2022/12/01/sta
nford-u-googles-convex-analytic-training-framework-improves-the-understanding-and-optimization-of-transformers/
… -
DeepMind Reduces Math Reasoning Errors Using Process-Based Supervision
By
–
DeepMind Studies Process- vs Outcome-based Model Supervision, Significantly Reducing Reasoning Errors on Math Word Problems https://
syncedreview.com/2022/11/30/dee
pmind-studies-process-vs-outcome-based-model-supervision-significantly-reducing-reasoning-errors-on-math-word-problems/
… -
Allen AI’s CLOSE Completes Visual Tasks From Text Inputs Alone
By
–
No Images Are Needed! Allen AI’s CLOSE Learns to Complete Visual Tasks From Text Inputs Alone