Top Tweets for #LLMBenchmarks

about 1 year ago

Your LLM evals might be burning cash for no reason. More evaluations ≠ better results. Generic metrics, excessive scope, and inadequate sampling are undermining your ROI. Smart judges need context, justification, and human validation. #AI #LLMBenchmarks #AIObservability

262

Jacobarrio @jlee8648

about 1 year ago

CURIE introduced custom evals like LLMSim and LMScore to grade nuanced outputs (like equations, summaries, YAML, code). Even the best models (Claude 3, Gemini, GPT-4) scored just ~32%. Proteins? Total fail. LLMs can read papers — solving them is another matter. #LLMbenchmarks #ArtificialInteligence #Google

WinBuzzer @WBuzzer

over 1 year ago

Tencent Releases its Hunyuan T1 AI Reasoning Model, Beating DeepSeek R1, GPT-4.5, o1 Across Multiple Benchmarks #AI #GenAI #TencentAI #HunyuanT1 #AIReasoning #EnterpriseAI #LLMbenchmarks #ChinaAI #MMLU #MathAI #AIModels #AIInference https://t.co/UgClFbhQj8

219

prasad kompalli @pkompalli

over 1 year ago

A small observation : more than solving HL math/physics/coding problems, I find getting LLMs to 'formulate' good set of solvable problems in a given topic ( algebra, geometry ... ) is a challenge. LLMs should be benchmarked in this. #GenAI #LLMbenchmarks

159

Data Science Dojo

@DataScienceDojo

over 1 year ago

Evaluating Your LLM? Here’s the Secret Sauce to Get it Right! 📊 Dive into the key metrics and methods that can help you assess and fine-tune your large language model, so it’s ready for the real world. https://t.co/XSVxf4mxlH #LLMs #LLMEvaluation #LLMBenchmarks

Simon P

@simonkp

over 2 years ago

@emollick While the giant context window and video capabilities grab headlines, Gemini Pro 1.5's core model performance shouldn't be overlooked. Surpassing Ultra 1.0 and nearing GPT-4 is impressive. Eager to see how this translates to real-world applications! #LLMBenchmarks #AIInnovation

418

Top Tweets for #LLMBenchmarks

Last Seen Hashtags on Sotwe

Trends for you

Most Popular Users