Inseq @InseqLib
Open-Source Interpretability for Generative Language Models 🔎 🐛 github.com/inseq-team/ins… Europe Joined November 2022-
Tweets57
-
Followers340
-
Following609
-
Likes94
Model Internals-based RAG Evaluation (MIRAGE) 🌴 is accepted to #EMNLP2024 Main! ➡️ To celebrate, here's our new MIRAGE demo combining @InseqLib and Transformers-specific LRP: huggingface.co/spaces/gsarti/…. Reach out if you want to catch up in Miami! 🤗🏖️
[1/8] Struggling with verifying the trustworthiness of RAG outputs? Check our latest work where we utilize *model internals* as a powerful and faithful tool for attributing answers to retrieved docs! (w/ @gsarti_ @AriannaBisazza @raquel_dmg) 📄: arxiv.org/abs/2406.13663 #NLProc
Very hyped for the new beautiful viz that just landed in the @InseqLib main branch! 🔥 This will empower users to explore attribution tensors more flexibly and intuitively. h/t to @_ddjohnson for his awesome work on the treescope toolkit powering this release!
Thanks to the new treescope integration, @InseqLib now supports interactive visualizations for multidimensional attributions (show_granular), token highlights (show_tokens) and improved viz for attribute_context CLI! 🚀 Install main, will appear in v0.7 x.com/_ddjohnson/sta…
Thanks to the new treescope integration, @InseqLib now supports interactive visualizations for multidimensional attributions (show_granular), token highlights (show_tokens) and improved viz for attribute_context CLI! 🚀 Install main, will appear in v0.7 x.com/_ddjohnson/sta…
By popular demand, the Treescope pretty-printer from the Penzai neural net library can now be installed separately, and supports both JAX and PyTorch! And that's not all: Penzai itself now has less boilerplate and includes more pretrained Transformer models!
🎓 We're thrilled to host Gabriele Sarti (@gsarti_) in our PhD seminar series tomorrow, July 16th, from 12:00-13:00 in Oe67 BU 101! Join us for his talk on interpreting context usage in generative language models, featuring the Inseq toolkit and PECoRe framework. 🕛Don't miss it!
The 🐑 PECoRe / 🌴 MIRAGE demo on @huggingface Spaces is powered by our new attribute-context CLI command released in v0.6, and allows to export the code to reproduce your results locally with 🐛 Inseq. Check it out ➡️ hf.co/spaces/gsarti/…
⚠️ Citations from prompting or NLI seem plausible, but may not faithfully reflect LLM reasoning. 🏝️ MIRAGE detects context dependence in generations via model internals, producing granular and faithful RAG citations. 🚀 Demo: huggingface.co/spaces/gsarti/… Fun collab w/ @Jirui_Qi,
Today, we had the first seminar of our #XAI course! @gsarti_ presented the @InseqLib to interpret LMs and the PECORE framework to identify & attribute context dependence in LMs! 🚀🌟 Thank you, it was so interesting! 🤗 Great start to our series! gsarti.com/talk/polito-in…
[1/4] Introducing “A Primer on the Inner Workings of Transformer-based Language Models”, a comprehensive survey on interpretability methods and the findings into the functioning of language models they have led to. ArXiv: arxiv.org/pdf/2405.00208
@MorenoLaQuatra And you haven't found the easter egg yet! 😇
Today @InseqLib hit 300 ⭐️ on Github! A huge thank you to all our awesome users ❤️ Onwards to the next 300! 🤺
@InseqLib v0.6 is out now on PyPI! 🔥 New CLI command for context attribution (@gsarti_), new perturbation-based methods by @hmohebbi75 & @casszzx and optimizations incl. multi-gpu support! ⚡️ Huge shoutout to our contributors! ❤️ Release notes ⬇️ github.com/inseq-team/ins…
Want to learn how to detect and attribute context usage in LMs using @InseqLib? This @Gradio demo using our CLI can teach you how! 🐛
The official 🐑 PECoRe 🐑 demo to detect and attribute context dependence in LM generations is now available on @huggingface Spaces! 🚀 Includes code examples, a usage guide, useful presets for various dec-only & enc-dec models, and more! Check it out ⬇️ huggingface.co/spaces/gsarti/…
🐑 PECoRe repository is now public (github.com/gsarti/pecore) and all model/datasets are available on @huggingface (huggingface.co/collections/gs…)! 🐑Interested in using PECoRe on your models? Have a look at the @InseqLib implementation (`inseq attribute-context`)!
[1/8] Our new work (w/ @AriannaBisazza @gchrupala @MalvinaNissim) is finally out! 🎉 We introduce PECoRe, an interpretability framework using model internals to identify & attribute context dependence in language models. 📄Paper: arxiv.org/abs/2310.01188 #NLProc #neuralempty
Tracker for Mamba gradient-based methods support: github.com/huggingface/tr…
Did someone say Mamba support? 👀 (Actually gradient-based methods are not working out-of-the-box because of some in-place variable overwriting, but should be fixable!)
`mamba` is now available in transformers. PEFT finetuning example: gist.github.com/ArthurZucker/7… Thanks @_albertgu and @tri_dao for this brilliant model! 🚀 and the amazing `mamba-ssm` kernels powering this!
Excited to present my recent work on @InseqLib and the #ICLR2024 PECoRe interpretability framework for the @SheffieldNLP group this afternoon! Many thanks @casszzx for inviting me! 🤗 🐛 Inseq: github.com/inseq-team/ins… 🐑🐑 PECoRe: openreview.net/forum?id=XTHfN…
Wow!! This is great news🥳 Thanks @InseqLib! I’m excited to try it out!
Value Zeroing, a faithful approach for analyzing context mixing in Transformers, is now available on @InseqLib main branch for all @huggingface text generation models! 🔀 🔍Paper introducing VZ: aclanthology.org/2023.eacl-main… 🐛VZ in Inseq: tinyurl.com/inseq-vz
@huggingface Authors: @hmohebbi75 @wzuidema @gchrupala @afraalishahi
Value Zeroing, a faithful approach for analyzing context mixing in Transformers, is now available on @InseqLib main branch for all @huggingface text generation models! 🔀 🔍Paper introducing VZ: aclanthology.org/2023.eacl-main… 🐛VZ in Inseq: tinyurl.com/inseq-vz
After two years with Poetry, Inseq just moved to @astral_sh's new blazing fast package manager uv! Our CI installation step is now ~80% faster! 🔥 Congrats to @charliermarsh and the team on the release, and godspeed for the cargo-like experience you are planning to build!
Interested in using @InseqLib for your work? Refer to our new tutorial to get started ➡️ github.com/inseq-team/ins… 8/8
Wang et al. propose a metric for LMs factual reliability and show its relation to models’ sensitivity to in-context distractors arxiv.org/abs/2310.09820 (W. Wang @bazril @alexandrabirch1 W. Peng @EdinburghNLP) 7/
Mousavi et al. attribute the generation of dialogue models, finding increased influence for more refined versions of dialogue history aclanthology.org/2023.nlp4conva… (@mahedmousavi S. Caldarella G. Riccardi @sislab7) 3/
akif @emremehmetakif
0 Followers 75 Following
eason_hsu @X0eY2VyXAJ48171
14 Followers 16 Following
Ceyhun Efe Kayan @cekayan
71 Followers 581 Following PhD Student @ Drexel CS | Mechanistic Interpretability
prasad @Mahadikprasad15
1K Followers 4K Following Just Curious. Prev crypto @Delphi_Digital @mhventures, Mech Interp
Souvik Bhattacharya @souvikb1812
87 Followers 5K Following
Byzantine* @KwesiEk
156 Followers 8K Following electrical engineer by academics and a web3 mechanic and a crypto trader . i believe in the power of Decentralize technologies and DAOs prog Language: Rust Dev
Sachit Malik @isachitmalik
165 Followers 4K Following Hola | Security Engineering at Apple | Alum: Carnegie Mellon; IIT Delhi
Yoram Bachrach @yorambac
4K Followers 7K Following Research Scientist at Meta (prev Google DeepMind and Microsoft Research). Working on LLM Agents and Multi-Agent Systems.
Gpbhupinder @gpbhupinder
385 Followers 8K Following 🚀Generative AI | Python | Deep Learning | React, Next.js 👨💻 Full-Stack Developer & AI Integration Expert
Jean Pascal @JeanPasqual99
146 Followers 8K Following
FBK - NLP Research Gr... @fbk_nlp
56 Followers 333 Following #NLP Research Unit at @fbk_research | Site: https://t.co/CrTKHvVqzI Bluesky: https://t.co/pPzvu4l4to #FBKresearch
David Ng'ang'a @dewkenzz
755 Followers 1K Following Atmospheric scientist focusing on the nexus between air quality, climate change and human health. Views expressed are mine!
Mike Allton | AI Stra... @mike_allton
50K Followers 8K Following Director @Agorapulse. I help Sales & Marketing leaders build "Digital Crews" to reclaim 70% of lost capacity. Speaker | Author | Practitioner. 🎙️ The AI Hat
Donatella Genovese @d_genovese
40 Followers 141 Following PhD Student at @SapienzaRoma | Works on Explainable AI at @IspammL Lab | prev. visit at LEAP Lab @ETH_en | https://t.co/rz4ueYDuoU
Giulia @giulia76330622
4 Followers 106 Following
Tanise Ceron @taniseceron
257 Followers 405 Following Postdoc @MilaNLProc | Interested in language models and how they shape the information environment
Gurumurthi V Ramanan @SuraSys1
740 Followers 7K Following CEO / MD, SuraSys - Building ML Systems for 19+ years
Laura King @KingLaura53775
126 Followers 4K Following
Sara Candussio @SaraCandussio
13 Followers 46 Following Applied Data Science and Artificial Intelligence PhD student @UniTrieste
chunhualiu @chunhualiu596
4 Followers 435 Following
Sabir Abdallah @sahirsabir35
6 Followers 150 Following
Oliver Eberle @eberle... @EberleOliver
292 Followers 816 Following Senior Researcher Machine Learning @TUBerlin @bifoldberlin 🇩🇪 Prev: @ipam_ucla | @UCLA | BCCN Interpretability | XAI | ML for Science
Vincent Weisser @vincentweisser
29K Followers 6K Following ceo @primeintellect — open stack for self-improving agents
Nathanael Carraz @carraznathanael
44 Followers 174 Following PhD student at Universitat Pompeu Fabra
Ashutosh Rudraksh @Ashu1190781
78 Followers 978 Following AI Engineer @meta | Co-founder @NavOn | MSc CS @ OSU | Building LLMs, HPC I/O & scalable systems | Tech-Podcast | Learning Crypto
Leesos @LeesosWoVOGZ8
50 Followers 4K Following
Jinyeong Kim @rubatoyeong
8 Followers 79 Following
Francois Meyer @francoisrmeyer
376 Followers 442 Following Lecturer at the University of Cape Town, working on low-resource NLP, data-efficient modelling, and linguistically informed interpretability.
Arturo Villacañas @arturovllacanas
214 Followers 1K Following Empirical research on AI learning offensive security. Ex @CambridgeMLG, @CISPA, @IMDEA_Software, @CCNCERT. rt ≠ endorse, opinions mine.
Daniel Johnson @_ddjohnson
3K Followers 970 Following Member of Technical Staff at @TransluceAI. Building tools to study neural nets and their behaviors. He/him.
belerico @belerico1
23 Followers 495 Following
Tommaso Tosato @Tommaso_Tosato
274 Followers 879 Following Postdoc at Mila and UdeM. Studying language in humans and machines. Focusing now on AI Safety.
Anastasiia Sedova @anasedova
140 Followers 354 Following Postdoc @NLPatITU ML Researcher @Apple MLR | #NLProc | prev: @CisLMU, @ViennaNLP, alumna @DAAD_Germany
ELOQUENCE AI @eloquenceai
299 Followers 2K Following #Multilingual and Cross-cultural interactions for context-aware, and #bias-controlled dialogue systems for safety-critical applications
Pol Pastells @PolPastells
26 Followers 690 Following Enginyer/Investigador PLN | Físic | 70% Lingüista NLP Research Engineer | Physicist | 70% Linguist
EP @EP225654
204 Followers 8K Following
Angelo Basile @angelo_basile
244 Followers 692 Following Research Scientist @SymantoAI. Doing things with words.
Boltzmann Labs @LabsBoltzmann
397 Followers 1K Following A Company with the vision of revolutionizing Drug Discovery. Designing precision medicine with data-driven solutions.
Arkajyoti Chakraborty @_arko_rico
4 Followers 204 Following Budding NLP Researcher :) 1 - AAAI'22, LREC-COLING'24!
Ruizhe Li @liruizhe94
1K Followers 2K Following Lecturer (Assistant Professor) @ABDNCompSci | Postdoc research fellow @ucl_wi_group | PhD CS @SheffieldNLP | mechanistic interpretability, multimodal LLMs
Álvaro Martinez @zenitramoravla
0 Followers 393 Following
Nastya Shaboltas @n_shaboltas
50 Followers 98 Following PhD student @univgroningen, @GroNLP https://t.co/Cc6pzcl3Ay
Bion @flesheatingemu
501 Followers 7K Following Futurist philosophy, molec neuro/immuno, pathophys, software eng, AI enjoyer Made an Apache/MIT `tree` util with tokens, lines, and module components
Sakana AI @SakanaAILabs
73K Followers 0 Following Sakana AI is an AI R&D company based in Tokyo. Try Sakana Marlin, Sakana Chat → https://t.co/1m2lSgnfB2
Daniel Johnson @_ddjohnson
3K Followers 970 Following Member of Technical Staff at @TransluceAI. Building tools to study neural nets and their behaviors. He/him.
NLP @ Uni Vienna @ViennaNLP
157 Followers 94 Following Digital Philology group led by Prof. Ben Roth Data Mining and Machine Learning research group @csunivie @univienna #nlpunivie #NLProc
ELOQUENCE AI @eloquenceai
299 Followers 2K Following #Multilingual and Cross-cultural interactions for context-aware, and #bias-controlled dialogue systems for safety-critical applications
Ruizhe Li @liruizhe94
1K Followers 2K Following Lecturer (Assistant Professor) @ABDNCompSci | Postdoc research fellow @ucl_wi_group | PhD CS @SheffieldNLP | mechanistic interpretability, multimodal LLMs
Nastya Shaboltas @n_shaboltas
50 Followers 98 Following PhD student @univgroningen, @GroNLP https://t.co/Cc6pzcl3Ay
Simone Tedeschi @SimoneTedeschi_
996 Followers 1K Following Applied Scientist @Amazon AGI • PhD @SapienzaRoma
Ziyang Luo @ChiYeung_Law
2K Followers 3K Following Working on Agentic RL @SFResearch | WizardCoder Creator | Optimizing for the long-term reward.
Fanny Jourdan @Fannyjrd_
2K Followers 264 Following French mathematician doing XAI & Fairness for #NLProc at @Mila_Quebec & @irtSaintEx. Prev: PhD on CS at @ANITI_Toulouse & MSc on maths at @polytechnique.
Raphaël Millière @raphaelmilliere
11K Followers 3K Following AI & Cognitive Science @UniofOxford @EthicsInAI Fellow @JesusOxford @raphaelmilliere.com on 🦋 Blog: https://t.co/2hJjfShFfr
Badr M. Abdullah, PhD... @badr_nlp
972 Followers 2K Following Researcher @LSTSaar | Saarland University🦉 💬 Speech & Language Processing 💫 Machine Learning 🤖 Cognitive Science 🧠
Lukas Galke Poech @LukasGalke
626 Followers 1K Following Assistant Professor @UniSouthDenmark | PI MIST Project on Scalable Mechanistic Interpretability for Safe & Trustworthy LLM Agents
Mohsen Fayyaz @mohsen_fayyaz
293 Followers 488 Following CS PhD Student @ UCLA #NLProc #MachineLearning
Eliana Pastor @eliana__pastor
356 Followers 510 Following Assistant Professor at PoliTo - @dbdmg_ she/her | #TrustworthyAI #XAI #FairnessAI
Philip Tannor @PhilipTannor
6K Followers 5K Following CEO at Deepchecks | Moderator at https://t.co/eIctpd8n3A | Forbes 30 Under 30 | Open Source Validation of AI & LLMs https://t.co/e8ivMRLuEp
António Correia @t0zecorreia
595 Followers 4K Following Ph.D. in Computer Science, Postdoctoral Researcher and member of the teaching staff (equivalent to Assistant Professor) @uniofjyvaskyla (formerly @MSFTResearch)
Daniel Dobriy (Q11227... @semantic42
2K Followers 2K Following Knowledge Graphs Researcher @wuvienna & Programme Manager @KMA_HQ | #digitalhumanism #knowledgegraphs #ai | https://t.co/xoERSGOyvN
AmsterdamNLP @AmsterdamNLP
4K Followers 325 Following Tweeting about NLP research, events and opportunities in Amsterdam -- run by @wzuidema and others.
Jirui Qi @Jirui_Qi
220 Followers 152 Following Ph.D Candidate @GroNLP, University of Groningen. RL + Multilingual + NLP = Future of world-wide #AI
Matthias Orlikowski @morlikow
611 Followers 2K Following NLProc, Computational Social Science • Human Label Variation, Disagreement, Subjectivity • PhD candidate @unibielefeld • he/him • EN, DE
Alina Leidinger @AlinaLeidinger
390 Followers 606 Following PhD student in NLP+AI Ethics @UvA_Amsterdam || prev mathematics @imperialcollege @TU_Muenchen she/her
Ana-Maria Bucur @bucuram
1K Followers 768 Following Postdoctoral Researcher at IDSIA @USI_INF. PhD @PRHLT & @UNIBUC. Working on NLP for Mental Health 🧠
Yuntian Deng @yuntiandeng
10K Followers 4K Following language models, world models | AP @UWaterloo | Associate @Harvard | Faculty Affiliate @VectorInst | Former Postdoc @ai2_mosaic | PhD @Harvard
Leonidas Zotos @ZotosLeo
4 Followers 2 Following
Daniel Scalena @daniel_sc4
205 Followers 804 Following Intern of TS @cohere | PhDing @unimib 🇮🇹 & @GroNlp 🇳🇱, interpretability et similia
Martijn Bartelds @BarteldsMartijn
657 Followers 425 Following Researcher @togethercompute | Formerly @stanfordnlp, @univgroningen, @tudelft and @Penn
sheoguo @sheo63009008
73 Followers 999 Following In my gap year. A football fun⚽️; A guitar player🎸.
Harish Tayyar Madabus... @harish
2K Followers 2K Following Senior Lecturer (Associate Professor) in Artificial Intelligence. Work on NLP (#NLProc) and Large Language Models
Luuk @highfrontvowel
256 Followers 431 Following Survivor of the Hobbesian state of nature | NLP Engineer
Hamidreza Amirzadeh @Hamiiiidreza99
13 Followers 132 Following AI Student @ Sharif University of Technology | Researching on NLP interpretability
Michael Hanna @michaelwhanna
822 Followers 506 Following PhD student at the University of Amsterdam / ILLC, interested in computational linguistics and (mechanistic) interpretability.
Gaofei Shen @linguisticshen
90 Followers 586 Following he/they, scienting computational linguist, photographer, ex radiation worker
Understandable Machin... @UMI_Lab_AI
646 Followers 123 Following Understandable Machine Intelligence Lab: We bring #explainable #AI to the next level. Part of @LeibnizATB, Ex @TUBerlin, funded by @BMBF_Bund #XAI
jacopo staiano @stjaco
1K Followers 5K Following *twitter won't let me follow you back so I put you in a private list*
Cas (Stephen Casper) @StephenLCasper
7K Followers 4K Following Computer scientist working on AI safeguards and gov research. Incoming assistant professor @Kennedy_School @Harvard. https://t.co/r76TGxSVMb
Lilian Weng @lilianweng
259K Followers 178 Following Co-founder of Thinking Machines Lab @thinkymachines; Ex-VP, AI Safety & robotics, applied research @OpenAI; Author of Lil'Log
Arena.ai @arena
167K Followers 215 Following Where AI meets the real world. Formerly LMArena. We measure and advance the frontier of AI through community-driven evaluation. We’re hiring → https://t.co/XBZCrseaWF
Chiara Di Bonaventura @c_dibonaventura
218 Followers 578 Following 👩💻 PhD student @KingsCollegeLon & @imperialcollege • #NLProc, LLMs, AI Safety • currently @turinginst • previously @Unibocconi @dartmouth
Raphael Pisoni @ml_4rtemi5
570 Followers 400 Following I love science, technology and engineering, I'm fascinated by the speed and power of innovation and i can't wait for the future! ML Researcher
Michela Lorandi @MichelaLorandi
52 Followers 98 Following PhD student at Dublin City University, ADAPT centre and SFI Centre for Research Training in Digitally Enhanced Reality (d-real)
MT Group at FBK @fbk_mt
1K Followers 455 Following #MachineTranslation Research Unit @FBK_research. #nlproc #deeplearning #ai
Madhumita Sushil @madhumitasushil
538 Followers 841 Following Assistant Professor @UCSF_DOCIT, @NeurosurgUCSF, @UCSF_BCHSI. Prev clinical NLP, neural network interpretability @clipsua, @Google Research
Uri Katz @UrikaUri
396 Followers 741 Following NLP PhD student @biunlp. Working on IR and Search, QA Reasoning. Intern at @allen_ai 🍄🟫🌵🍄🍄🟫🌵🍄🍄🟫🌵🍄
Vivek Gupta @keviv9
4K Followers 6K Following Assistant Prof @SCAI_ASU; PostDoc @cogcomp @Penn, ed-@UUtah,@iitkanpur. @Bloomberg @MSFTResearch Fellow; ex-@MetaAI @IBM @samsungresearch
Luca Dini @lucadini_
42 Followers 126 Following PhD student @ italiaNLP_lab (ILC-CNR) NLP • Deep Learning #NLProc
Laura Occhipinti @LauraOcchipint8
35 Followers 341 FollowingYou might like
