https://
huggingface.co/spaces/openbmb
/MiniCPM-V-2_6
…
LLMS
-
MiniCPM-V-2_6 Model Space on Hugging Face
By
–
-
Anonymous GPT-4 based model performance on LMSYS Arena
By
–
A new Anonymous GPT-4 based model seems to be better at writing a bit (responses are more humanised). But I had the same impression from Google’s Eureka (which I am still not sure if it was already released) ICYMI: you can spot it on lmsys arena in the battle mode
-
Dangers of LLM-Generated Training Data
By
–
One of the worst mistakes many AI teams are making today: Using data generated by LLMs to train machine learning models or create a 'self-learning' training loop. While the idea may seem appealing and
-
GPT-4’s RL vs Imitation: A Student Playing a Teacher’s Quirky Grading System
By
–
This was very noticeable in GPT-4 around its release and anecdotally it’s gotten better since. My complete armchair guess is this is RL rather than imitation — it feels like a student playing to their teacher’s quirky grading system.
-
LLaVA-OneVision: Open Large Multimodal Models for Visual Task Transfer
By
–
LLaVA-OneVision Easy Visual Task Transfer discuss: https://
huggingface.co/papers/2408.03
326
… We present LLaVA-OneVision, a family of open large multimodal models (LMMs) developed by consolidating our insights into data, models, and visual representations in the LLaVA-NeXT blog series. Our -
Google Announces CoverBench for Complex Claim Verification
By
–
Google announces CoverBench A Challenging Benchmark for Complex Claim Verification discuss: https://
huggingface.co/papers/2408.03
325
… There is a growing line of research on verifying the correctness of language models' outputs. At the same time, LMs are being used to tackle complex queries that -
Google: Test-Time Compute Scaling More Effective Than Model Parameters
By
–
Google announces Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters discuss: https://
huggingface.co/papers/2408.03
314
… Enabling LLMs to improve their outputs by using more test-time computation is a critical step towards building generally self-improving -
MMIU: Evaluating Large Vision-Language Models with Multiple Images
By
–
MMIU Multimodal Multi-image Understanding for Evaluating Large Vision-Language Models discuss: https://
huggingface.co/papers/2408.02
718
… The capability to process multiple images is crucial for Large Vision-Language Models (LVLMs) to develop a more thorough and nuanced understanding of a scene.