msr_knowledgenlp

Microsoft Cognitive Services Research, Knowledge and Language team,

[AK @_akhaliq]

Any-to-Any Generation via Composable Diffusion

present Composable Diffusion (CoDi), a novel generative model capable of generating any combination of output modalities, such as language, image, video, or audio, from any combination of input modalities

paper page: https://t.co/Tj3X8UvjlQ

[Canwen Xu @XuCanwen]

Do you know small models can be LLM's plugins?
🥳Introducing SuperICL, it's ICL but super dope!
🚀SuperICL combines small models with LLM like GPT-3.5 and can improve accuracy, multilinguality and interpretability!
📄Paper: https://t.co/OCIB6lXqom
🔧Code: https://t.co/ZKzIdtWltg

[AK @_akhaliq]

Small Models are Valuable Plug-ins for Large Language Models

abs: https://t.co/NU7fnVcfov
paper page: https://t.co/lnxoGS1bds

[Meng Jiang @Meng_CS]

THE BATCH selected four pieces of news after today's Andrew's newsletter, from Hinton leaving G to @wyu_nd GenRead paper at ICLR. The work is open sourced. Wenhao is on the market! Grab him asap!

https://t.co/BG6FyYEU02

https://t.co/XvUgeVgiPo

[KnowledgeNLP Workshop @NAACL 2025 @knowledgenlp]

The second workshop on Knowledge Augmented Methods for NLP at #KDD2023 is welcoming submissions📷! Papers due by May 23rd! Accepted paper will be non-archival! Details are available 📷 https://t.co/NO28bVXQ1N

[AK @_akhaliq]

GPTEVAL: NLG Evaluation using GPT-4 with Better Human Alignment

abs: https://t.co/CRPk20t5zx

[msr_knowledgenlp @MS_KnowledgeNLP]

@LiGuangruii Feel free to reach out our researchers, https://t.co/3WnqCqxqu3

[msr_knowledgenlp @MS_KnowledgeNLP]

We are hiring part-time research interns on LLM. Please message us if you are interested : )

[msr_knowledgenlp @MS_KnowledgeNLP]

@SinclairWang1 @zhangzhuosheng Yes : )

[msr_knowledgenlp @MS_KnowledgeNLP]

Thank you for the interest in our team! Sorry for not being able to reply everyone. We will keep hiring interns quarter by quarter. Feel free to message us your personal page or CV by Twitter whenever you are available. Our researchers will talk with you if matched : )

[msr_knowledgenlp @MS_KnowledgeNLP]

@amelie_reymond Oh, sorry, just enabled it. Could you try it again? Thanks!

[msr_knowledgenlp @MS_KnowledgeNLP]

@amelie_reymond Message us your personal page or CV by Twitter Messages would be fine : )

[Wenhao Yu @wyu_nd]

Combing Retrieval AND Generation (in step1) can further improve the model performance, as shown in Figure 3. The choice of retrieval or generation is interesting, and their complementarity is worth exploring. Using retriever or generator only where it helps.

[KnowledgeNLP Workshop @NAACL 2025 @knowledgenlp]

🎉 Accepted papers (6 oral + 20 poster) are now online! Check them at https://t.co/bC8q67ibfm

[Wenhao Yu @wyu_nd]

𝗚𝗲𝗻𝗲𝗿𝗮𝘁𝗲 𝗥𝗮𝘁𝗵𝗲𝗿 𝗧𝗵𝗮𝗻 𝗥𝗲𝘁𝗿𝗶𝗲𝘃𝗲 is now 𝗮𝗰𝗰𝗲𝗽𝘁𝗲𝗱 to #𝗜𝗖𝗟𝗥𝟮𝟬𝟮𝟯 🎉🎉 Without using DPR/Google, it achieved SoTA on multiple open-domain QA and knowledge-intensive benchmarks! Work done
@ms_knowledgenlp!
Code and paper: https://t.co/hfzN9bAneC

[KnowledgeNLP Workshop @NAACL 2025 @knowledgenlp]

Call for papers! The first workshop on Knowledge Augmented Methods for NLP (#NLProc) at #AAAI2023 is welcoming submissions🙌! Papers due on Nov. 4! Papers will be non-archival, so published papers (e.g.#EMMLP2022) can also present at our workshop! Details👉https://t.co/Xu8VbGb3or