Shi💙

🧠 Neuroscience gave AI its start. What happens now? 🤖 Neuroscience helped launch AI, but David Sussillo, a computational neuroscientist, believes its next major contribution may take longer than many expect. Meanwhile, AI is becoming a powerful tool for studying the brain. Central points: 📍The fields have diverged: modern AI increasingly advances through engineering rather than biological insights. 📍Neuroscientists still have valuable expertise: their methods can help investigate AI’s internal workings and behavior. 📍Timescales matter: he doubts brain research will redirect AI within five years, but anticipates potentially important contributions over 10–20 years. 📍A possible bridge: large models built from high-quality neural recordings could enable experiments that are difficult in living brains. Note: This remains a proposed direction. By @TheTransmitter #Neuroscience #AI #Future 👉thetransmitter.org/artificia… @Khulood_Almani @Shi4Tech @jblefevre60 @dinisguarda @Eli_Krumova @BroadenView @HaroldSinnott @ThEiLe0 @Mack1510019 @drsharwood @EvanKirstel @ahier @RLDI_Lamy @Timothy_Hughes @Ym78200 @Nicochan33 @mvollmer1 @jeancayeux @rvp @SusanFourtane @amalmerzouk @bamitav @Fabriziobustama @FrRonconii @domingonarvaez1 @c4trends @TerenceLeungSF @aure79lien @EstelaMandela @NevilleGaunt @stanleywaite1 @pchamard @corixpartners #Brain #NeuroAI #Megaphysics Entwined histories: In the short term, AI may help neuroscience, but in the long term, neuroscience may still unveil secrets for AI. Illustration by Andreea Jurj

It is my understanding that, without context-gating, the human brain would “catastrophically forget” Now, from all my postings about neuroscience and learning about the brain’s plasticity, for me, this article beautifully highlights the fact that Al cannot think like a human, because it does not have dendritic-branch plasticity, which is the brain's fundamental mechanism for learning without forgetting. Dear Shi @Shi4Tech I very much appreciate your shared enthusiasm and support. Wishing you a very happy day ahead 🧠🙏

🤖 Better Answers, Better Confidence: A Step Toward Responsible AI — Enhancing AI Accuracy and Confidence in Answer Generation 🧠 The #ANSPRE study by #JAIST highlights several benefits: 📍More accurate answers: improved factual question-answering performance on the benchmarks tested. 📍Clearer responses: guides AI toward concise, specific answers. 📍More meaningful confidence scores: scores better reflect whether an answer is correct. 📍Better use of evidence: combines candidate answers from multiple retrieved documents. 📍 Flexible integration: can complement existing language models and retrieval systems. 🧠🤖 In essence, this research offers a promising step toward responsible AI by making answers more accurate and their confidence easier to assess, while still requiring verification. By @NeuroscienceNew #ResponsibleAI #Accuracy #ML #LLM 👉neurosciencenews.com/ai-llm-… @ipfconline1 @Khulood_Almani @Shi4Tech @dinisguarda @Eli_Krumova @BroadenView @ThEiLe0 @HaroldSinnott @jblefevre60 @Mack1510019 @drsharwood @EvanKirstel @ahier @RLDI_Lamy @Timothy_Hughes @Ym78200 @Nicochan33 @mvollmer1 @jeancayeux @SusanFourtane @amalmerzouk @bamitav @Fabriziobustama @FrRonconii @domingonarvaez1 @c4trends @TerenceLeungSF @aure79lien @EstelaMandela @NevilleGaunt @stanleywaite1 @pchamard @corixpartners @rvp #RAG #TechForGood #EnterpriseAI Description: Another important aspect of LLMs is their ability to produce confidence scores, which reflect how certain the model is about the correctness of its answer. Credit: Neuroscience News

🧠 How can AI learn without forgetting? 📍 A new computational model suggests that context-dependent control of learning within dendritic branches could help new memories form while preserving old ones — offering a possible explanation for how brains resist catastrophic forgetting. By @TheTransmitter #NeuroAI #DeepLearning #Neuroscience #AI 👉thetransmitter.org/systems-n… @Khulood_Almani @Shi4Tech @dinisguarda @Eli_Krumova @ThEiLe0 @HaroldSinnott @jblefevre60 @ipfconline1 @Mack1510019 @drsharwood @EvanKirstel @ahier @RLDI_Lamy @Timothy_Hughes @Ym78200 @Nicochan33 @mvollmer1 @jeancayeux @SusanFourtane @amalmerzouk @bamitav @Fabriziobustama @FrRonconii @domingonarvaez1 @c4trends @TerenceLeungSF @aure79lien @EstelaMandela @NevilleGaunt @stanleywaite1 @pchamard @corixpartners @SwissCognitive #Brain #ANN #CatastrophicForgetting Description: Divided attention —different parts of the dendrite make distinct computations, preventing new memories from erasing old ones, according to a new study - Karya (istock)