Natural Language Processing A Machine Learning Perspective AUTHORS: Yue Zhang, Westlake University Zhiyang Teng, Westlake University DATE PUBLISHED: January 2021 ISBN: 9781108420211

📢Excited to Share AVERITEC with amazing Michael Schlichtkrull Andreas Vlachos! It is the first AFC dataset to avoid context dependence, evidence insufficiency, and temporal leaks. LLMs still struggle with this challenging task. Welcome to test your models on this benchmark!

It has been exactly one year since the release of ChatGPT. How far are open-source LLMs? We provide an exhaustive review of open-source LLMs that claim to catch up with or surpass ChatGPT in various capabilities. paper🔗: arxiv.org/pdf/2311.16989… 🧵(1/5)

Non-autoregressive models advance document-level machine translation with impressive speedup. Delve into the opportunities and challenges of NAT on extended sequences: aclanthology.org/2023.findings-… #EMNLP2023

Mobile ALOHA's hardware is very capable. We brought it home yesterday and tried more tasks! It can: - do laundry👔👖 - self-charge⚡️ - use a vacuum - water plants🌳 - load and unload a dishwasher - use a coffee machine☕️ - obtain drinks from the fridge and open a beer🍺 - open

Excited to announce that Fast-DetectGPT made it to #ICLR2024 🎉 WestlakeNLP In the rebuttal phase, got an unfair 1 amidst 8866. Huge shoutout to Eric for the public support. Appreciate your sense of justice! ⚖️🔍 Paper: openreview.net/forum?id=Bpcgc…

I am seeking a research intern to collaborate on the development and application of large language models within real-world industry scenarios at ByteDance, Singapore. If you are interested, please drop me an email.

Figuring out which topic to work on is probably the most challenging task for deep learning researchers these days. I wrote a blog post to give you some ideas. Read it here: ofir.io/Tips-for-Findi…

I just published “No Training Needed, Fast-DetectGPT Boosts Text Detection Speed by 340 Times” link.medium.com/sozGRpDd9Hb

👏Excited to share PairS🌟: Simple yet effective/efficient LLM Evaluators for various NLG tasks! 1⃣ more thing: We provide a rigorous problem formulation and transitivity investigation for LLM Evaluators. Check it out on: 📖arxiv.org/abs/2403.16950 📷github.com/cambridgeltl/p…

One more intern position is available at ByteDance, Singapore. If you are interested, please drop me an email.

MMLongBench-Doc has been accepted to the #NeurIPS2024 D&B Track. VLMs process documents in a WYSIWYG manner, providing better answers than the OCR + LLM pipeline. With powerful open-source VLMs rapidly being developed, it's time to focus on VLMs for document understanding!

Prediction: within the next year there will be a pretty sharp transition of focus in AI from general user adoption to the ability to accelerate science and engineering. For the past two years it has been about user base and general adoption across the public. This is very

Life update: 🎉 I'm excited to share that I will be joining HKUST Guangzhou as an Assistant Professor in Spring 2025! I'm looking for multiple PhDs and interns who are passionate about exploring research questions related to knowledge and reasoning in the context of LLMs. 🤖

🚀 Excited to share the AI-generated text detection demo: huggingface.co/spaces/gshbao/…🌟, for our ICLR 2025 work: "Glimpse: Enabling White-Box Methods to Use Proprietary Models for Zero-Shot LLM-Generated Text Detection"! WestlakeNLP #AI #ICLR2025 #TextDetection #MachineLearning

Yesterday, we shared the details on our xLSTM 7B architecture. Now, let's go one level deeper🧑‍🔧 We introduce ⚡️Tiled Flash Linear Attention (TFLA), ⚡️ A new kernel algorithm for the mLSTM and other Linear Attention variants with Gating. We find TFLA is really fast! 🧵(1/11)

字节跳动 sg科研实习生招聘： 1: 纯科研，无业务压力，做多模态大模型和强化学习推理方向。 2: 无学历要求，只要动手能力强。 3: 做事踏实严谨不耍花招。 4: 新加坡(Preferred)、北京、上海、深圳、广州、澳大利亚均可。

📢 (1/16) Introducing PaTH 🛣️ — a RoPE-free contextualized position encoding scheme, built for stronger state tracking, better extrapolation, and hardware-efficient training. PaTH outperforms RoPE across short and long language modeling benchmarks arxiv.org/abs/2505.16381

QK-Clip: Taking Muon Further on the Scaleup Journey kexue.fm/archives/11126 Interpreting the Key Training Techniques Behind Kimi K2: QK-Clip and MuonClip.

Best poster design I’ve seen at #acl2017nlp by Rob van der Goot