I think a very useful thing for AI critics to do would be to "red team" their own beliefs by spending some serious time and effort (maybe with help) to see if they can get AI to do the things they say it can't. If they are right, that is very valuable benchmarking information.
GENERATIVE AI
-
AI Writing Quality Compared to Human Writing Standards
By
–
Agree on both counts. But this was also a really simple prompt without any attempt to do multiple interactions, etc. AI writing is not near as good as good human writing, but it does beat that of a large number of people, especially when they haven’t been taught to write well.
-
AI Transforms Garment Industry: From Local Tailors to Designer Automation
By
–
Local darzi —-> readymade garments —-> designer darzi
-
Greatest Misconception: LLMs Complexity and Usability
By
–
That's my general view on all of this stuff now: the greatest misconception in LLMs is that the resulting tools are easy to use and to figure out
-
Copilot Training and Documentation Effectiveness in Organizations
By
–
Did your company spend any time teaching people how to use it effectively? (If they did I'd love to hear details – training and documentation on Copilot is incredibly rare in my experience – which is bad because the tool is very unintuitive to use well)
-
AI as Learning Tool, Not Job Shortcut for Programming
By
–
Yeah, anyone who's telling people they can get a job low without actually learning to program is doing nobody any favors It's a learning accelerant tool, not a replacement for learning
-
Aligning LLMs to Quote from Pre-Training Data
By
–
9/ Aligning LLMs to Quote from Pre-Training Data – proposes techniques to align LLMs to leverage memorized information quotes directly from pre-training data.
-
Iterative Self-Revision and Search for LLM Reasoning Tasks
By
–
6/ Reasoning with Intermediate Revision and Search – presents an approach for general reasoning and search on tasks that can be decomposed into components; incorporates iterative self-revision capabilities and allows an LLM to build an interwoven network of thoughts.
-
LM-Guided Chain-of-Thought Knowledge Distillation Technique
By
–
4/ LM-Guided Chain-of-Thought – applies knowledge distillation to a small LM with rationales generated by the large LM; the rationale is then generated by the lightweight LM and the answer prediction is done by the frozen large LM.
-
CodeGemma: Open Code LLM Family Matching Performance Benchmarks
By
–
3/ CodeGemma – a family of open code LLMs based on Gemma; CodeGemma 7B models excel in mathematical reasoning and match the code capabilities of other open models.