iglu-contest @ NeurIPS

Join us for another SIGIR Lean and Learn series on LLM-based Applications with "AutoGen: A Programming Framework for Agentic AI". P.s : I was honoured to work on AgentEval as part of Autogen as of my internship project @MSFTResearch. bit.ly/4eieaXz #LLMs #SIGIR2024 #LLM

🚀 Thrilled to see the #IGLU dataset showcasing the incredible potential of #LLMs in the dynamic world of Minecraft! 🌍 🏆🔍 #AI #MachineLearning #GamingInnovation #MinecraftAI arxiv.org/abs/2402.08392

Had an amazing time discussing #AutoGen and #AgentEval at the Bosch Center for Artificial Intelligence (BCAI) @Bosch_AI! 🚀Thanks to everyone who participated linkedin.com/feed/update/ur… #GenAI #Innovation #Community #research #AI @qingyun_wu @Chi_Wang_ @pyautogen

🌟Excited about #LLMs in search, recommendations, and much more? Join us at LLM Day @ACMSIGIR on July 16 in Washington, DC! We have prepared an exciting program for you! #SIGIR2024 📅 Schedule & Highlights: 1️⃣ 09:00 - 10:00 Keynote by Thorsten Joachims (Cornell University) 2️⃣ 10:30 - 11:00 Jason Weston @jaseweston (Meta) gives a talk about Self-Rewarding LLMs 3️⃣ 11:00 - 11:45 State of LLMs in Search and Recommendations: Chenyan Xiong @XiongChenyan (CMU): The Role of Information Retrieval in the Era of LLMs Xinyu Crystina Zhang (UWaterloo): Multilinguality in the Era of LLMs 4️⃣ 11:45 - 12:30 LLMs at Industry Scale: Dawei Yin (Baidu): LLMs in Baidu Search Amanda Bertsch @abertsch72 (CMU): Unlimiformer and Long Context Models 5️⃣ 13:30 - 15:00 Health, Legal, Finance, and Ethics: Subhabrata Mukherjee (Hippocratic AI): Safety-focused Conversational AI for Healthcare Michael Cole (LexisNexis): LLMs in the Legal Domain Anju Kambadur (Bloomberg): Generative AI in Finance Jordan Meyer (Spawning): AI Training Data and Consent 6️⃣ 15:30 - 16:30 Multi-Agents & Reasoning: Qingyun Wu (Penn State) @qingyun_wu & Chi Wang @Chi_Wang_ (Microsoft Research): #AutoGen Framework (@pyautogen) Omar Khattab @lateinteraction (Stanford): DSPy: Self-Improving Language Programs Mirac Suzgun (Stanford): Meta-prompting 7️⃣ 16:30 - 17:30 Anticipated Trends in Search: Benjamin Piwowarski (CNRS): Index-based Retrievers in the Neural Era Bryan McCann (from @yousearchengine): The Future of Trust in LLMs The full program is available here: sigir-2024.github.io/call_for_parti…

Do you remember when you joined X? I do! #MyXAnniversary

Latest research work by my colleague Tobias Schnabel: Supercharging Prompt Engineering via Symbolic Program Search. 🚀 Exciting stuff!!! towardsdatascience.com/supercharging-…

StateFlow is recently integrated into @pyautogen . It already helped some users achieving both higher quality and lower cost with their multi-agent system. Kudos to @YiranWu18 @VyvyenYue @ShaokunZhang1 @qingyun_wu for the timely research and fast integration.

New use case of @pyautogen in blockchain:

🎉 Exciting event alert! 🎉 Join us to discover the real-world applications of #AutoGen and how it's shaping the future! Don't miss out - mark your calendars! #innovation

#ClariQ competition co-organized at #SCAI2020 was an original inspiration for IGLU 😉 Collected dataset is still widely used! Everything has its purpose and place in the timeline✨ aclanthology.org/2021.emnlp-mai… convai.io w/@maliannejadi @varphi @JeffD @MikhailBurtsev

It was great to present our work done on top of @IgluContest dataset at #EACL2024! Kudos goes to @nikhilhockey for all the hard work! Great audience and a lot of nice discussions! Amazing organisation! Hope to see you all soon again!

Really nice poster on retrieving the cause of questions in the educational domain; I can see how it can improve lecture preparation and more #EACL2024 #eacl

Yet another cool idea and great poster design! #EACL2024

Checkout this new research on AI agents enabled by #AutoGen!

Excited to dive into Interactive Grounded Language Understanding? 🚀 Seeking to develop agents adept at natural dialogue, assisting in real-world tasks. Key: knowing when to ask for feedback and learning. Check out datasets to get started: github.com/microsoft/iglu… #EACL2024 #EACL

Stop by our poster @eaclmeeting on "Clarifying the Path to User Satisfaction: An Investigation into Clarification Usefulness". #EACL2024

Come stop by our poster @eaclmeeting #EACL2024

Introducing Orca-Math, our Mistral-7B offshoot excelling in math word problems! 🧮🐳 - Impressive 86.81% score on GSM8k - Surpasses models 10x larger or with 10x more training data - No code, verifiers, or ensembling tricks needed

After receiving a lot of help, and meeting the Maltese ambassador 😉 I am heading to #EACL, excited to share our work on interactive agents! Looking forward to catching up with old and new friends. DM me for a chat about anything, including @pyautogen arxiv.org/abs/2304.10750

For learning about task decomposition + #AutoGen, also see: microsoft.github.io/autogen/docs/t…