Join me to explore the future of technology with Vasudev Lal from Intel Labs, showcased at Dell Tech World. We discussed the innovative strides in Cognitive AI and the impact of the Gaudi platform on deep learning and industry solutions. #IntelAmbassador Watch the insightful
RESEARCH
-
Neural Operators Enable Resolution-Agnostic Function Space Mappings
By
–
Neural operators are an approach to generalize to new resolutions since they learn mappings between function spaces.
-
Clément Canonne Leads Sublinear Algorithms Boot Camp on Distribution Testing
By
–
The man, the myth, the legend: Clément Canonne (@ccanonne_) kicks off day 4 of the Sublinear algorithms boot camp with distribution testing! @simon #SimonsLive
-
Phi Models Overrated: Scaling Remains Safest Path Beyond GPT-4o
By
–
I'm slightly skeptical about Phi models, which are very good but overrated in benchmarks imo I'd say it's not necessary, but scaling is definitely the safest way of surpassing GPT-4o
-
Gemini Performance Improvement on MMLU Benchmark
By
–
Agreed, MMLU is not a great benchmark and Gemini has improved quite a lot
-
Clarifying Graph Vector Communication: Misleading vs Wrong Description
By
–
Ok I wouldn't say it was "wrong", just "misleading" :). The idea of having a vector stored at each node in the graph, and the vectors communicating via weighted sums over directed edges I think is all well and sound all by itself, at this level of description. It's misleading
-
Machine emotions: authenticity versus deception in constrained LLMs
By
–
Clear deception makes the interaction much less joyful. Similar to an LLM that pretends to feel real empathy. But this… the frustration or confusion expressed by this clamped model is… a new kind of machine-like feeling?
-
Major Tech Disruptions Expected in Next Five Months
By
–
Whatever you think of the last four years, it’s about to get much more insane and way more stupid in the next five months. Buckle up
-
Data Access and AI Collaboration Tools for Research
By
–
Very cool research subject! We would love to help by providing better data access in the future for this kind of research. We also dream of building tools that unlock magnitudes wider / more intensity of collaboration in AI. It's still the beginning