Lilian Weng
Machine-learning researcher; started the Lil'Log survey posts on how a model technique works in 2017. Worked at OpenAI from 2018 to 2024, latterly leading its safety systems team; on the founding team of Thinking Machines Lab from 2025 to July 2026, then rejoined OpenAI.
Lilian Weng did not write this page. ppll compiled it from their own words, each linked to the source. Is this you? · Report a mistake
Latest x.com
What they believe
-
The deployment layer surrounding a model is as important as the model's raw intelligence.
the layer between the raw model and the real-world context seems to be as important as the model's raw intelligence
lilianweng.github.ioSaid 4 Jul 2026This on korrents
-
As models keep growing, AI is running out of enough high-quality unique training tokens to keep up.
As the model size grows significantly, we are running out of enough high-quality unique tokens.
lilianweng.github.ioSaid 24 Jun 2026This on korrents
The bold line is korrents’ wording; the quote is theirs. These 2: what they wrote before what they said aloud, one per source, newest first.
Feed
As its own page →Hiding
27 July
15 July
14 July
7 July
4 July
From one piece Harness Engineering for Self-Improvement 3 beliefs · lilianweng.github.io
-
Their words
harness improvement enables better deployment of the model but intelligence is still the core.
Harness Engineering for Self-Improvementlilianweng.github.io
-
korrents.com
Recursive self-improvement only works if the base model is already capable enough to improve its own mechanism.Their words
Recursive structure alone is not enough. The base model must be capable enough to improve the mechanism.
Harness Engineering for Self-Improvementlilianweng.github.io
-
korrents.com
The deployment layer surrounding a model is as important as the model's raw intelligence.Their words
the layer between the raw model and the real-world context seems to be as important as the model's raw intelligence
Harness Engineering for Self-Improvementlilianweng.github.io
25 June
24 June
-
korrents.com
As models keep growing, AI is running out of enough high-quality unique training tokens to keep up.Their words
As the model size grows significantly, we are running out of enough high-quality unique tokens.
19 May
18 May
11 May
10 March
15 January
27 October 2025
1 October 2025
26 September 2025
10 September 2025
18 August 2025
1 May 2025
From one piece Why We Think 3 beliefs · lilianweng.github.io
-
Their words
model CoTs could be biased due to lack of explicit training objectives aimed at encouraging faithful reasoning
-
korrents.com
Longer test-time thinking improves an AI model's robustness to adversarial or unusual inputs.Their words
thinking for longer should be especially useful when the model is presented with an unusual input, such as an adversarial example or jailbreak attempt
-
korrents.com
Large language models cannot reliably self-correct their own mistakes without external feedback.Their words
this self-correction capability turns out to not exist intrinsically among LLMs and does not easily work out of the box, due to various failure modes
28 November 2024
From one piece Reward Hacking in Reinforcement Learning 2 beliefs · lilianweng.github.io
-
korrents.com
More capable AI agents are more likely to find and exploit flaws in their reward functions.Their words
A more intelligent agent is more capable of finding "holes" in the design of reward function and exploiting the task specification-in other words, achieving higher proxy rewards but lower true rewards.
Reward Hacking in Reinforcement Learninglilianweng.github.io
-
Their words
Reward hacking exists because RL environments are often imperfect, and it is fundamentally challenging to accurately specify a reward function.
Reward Hacking in Reinforcement Learninglilianweng.github.io
7 July 2024
From one piece Extrinsic Hallucinations in LLMs 2 beliefs · lilianweng.github.io
-
korrents.com
Using supervised fine-tuning to teach a language model new knowledge risks increasing its hallucination rate.Their words
These empirical results from Gekhman et al. (2024) point out the risk of using supervised fine-tuning for updating LLMs' knowledge.
-
korrents.com
A language model avoids hallucinating only by being factual and admitting when it does not know the answer.Their words
To avoid hallucination, LLMs need to be (1) factual and (2) acknowledge not knowing the answer when applicable.
21 May 2024
12 April 2024
From one piece Diffusion Models for Video Generation 2 beliefs · lilianweng.github.io
-
korrents.com
High-quality video data, especially paired with text, is harder to collect at scale than image or text data.Their words
In comparison to text or images, it is more difficult to collect large amounts of high-quality, high-dimensional video data, let along text-video pairs.
-
Their words
It has extra requirements on temporal consistency across frames in time, which naturally demands more world knowledge to be encoded into the model.
5 February 2024
-
korrents.com
High-quality data collection depends more on careful human execution than on machine learning techniques alone.Their words
Lots of ML techniques in the post can help with data quality, but fundamentally human data collection involves attention to details and careful execution.
25 October 2023
-
Their words
High perplexity makes an attack more vulnerable to be detected and mitigated.
1 September 2023
23 June 2023
25 March 2023
24 March 2023
15 March 2023
27 January 2023
10 January 2023
1 January 2023
8 September 2022
28 July 2022
9 June 2022
15 April 2022
22 March 2019
27 August 2018
14 October 2014
14 February 2014
Nothing matches.
Beliefs
Korrents What they believe 16 beliefs — each backed by an exact quote.
Each is a — compiled by korrents.com, not by them: the one-line wordings are korrents', the quotes are theirs.
Recent
The deployment layer surrounding a model is as important as the model's raw intelligence.
the layer between the raw model and the real-world context seems to be as important as the model's raw intelligence
Harness Engineering for Self-Improvement Said 4 Jul 2026
Recursive self-improvement only works if the base model is already capable enough to improve its own mechanism.
Recursive structure alone is not enough. The base model must be capable enough to improve the mechanism.
Harness Engineering for Self-Improvement Said 4 Jul 2026
Harness improvements help deploy a model better, but the model's underlying intelligence remains the core driver of capability.
harness improvement enables better deployment of the model but intelligence is still the core.
Harness Engineering for Self-Improvement Said 4 Jul 2026
Show 13 more
As models keep growing, AI is running out of enough high-quality unique training tokens to keep up.
As the model size grows significantly, we are running out of enough high-quality unique tokens.
Scaling Laws, Carefully Said 24 Jun 2026
Large language models cannot reliably self-correct their own mistakes without external feedback.
this self-correction capability turns out to not exist intrinsically among LLMs and does not easily work out of the box, due to various failure modes
Why We Think Said 1 May 2025
Longer test-time thinking improves an AI model's robustness to adversarial or unusual inputs.
thinking for longer should be especially useful when the model is presented with an unusual input, such as an adversarial example or jailbreak attempt
Why We Think Said 1 May 2025
Chain-of-thought explanations from language models can be biased because training does not explicitly reward faithful reasoning.
model CoTs could be biased due to lack of explicit training objectives aimed at encouraging faithful reasoning
Why We Think Said 1 May 2025
It is fundamentally difficult to design a reward function that accurately captures the intended goal in reinforcement learning.
Reward hacking exists because RL environments are often imperfect, and it is fundamentally challenging to accurately specify a reward function.
Reward Hacking in Reinforcement Learning Said 28 Nov 2024
More capable AI agents are more likely to find and exploit flaws in their reward functions.
A more intelligent agent is more capable of finding "holes" in the design of reward function and exploiting the task specification-in other words, achieving higher proxy rewards but lower true rewards.
Reward Hacking in Reinforcement Learning Said 28 Nov 2024
A language model avoids hallucinating only by being factual and admitting when it does not know the answer.
To avoid hallucination, LLMs need to be (1) factual and (2) acknowledge not knowing the answer when applicable.
Extrinsic Hallucinations in LLMs Said 7 Jul 2024
Using supervised fine-tuning to teach a language model new knowledge risks increasing its hallucination rate.
These empirical results from Gekhman et al. (2024) point out the risk of using supervised fine-tuning for updating LLMs' knowledge.
Extrinsic Hallucinations in LLMs Said 7 Jul 2024
Video generation requires more world knowledge than image generation because of the need for temporal consistency across frames.
It has extra requirements on temporal consistency across frames in time, which naturally demands more world knowledge to be encoded into the model.
Diffusion Models for Video Generation Said 12 Apr 2024
High-quality video data, especially paired with text, is harder to collect at scale than image or text data.
In comparison to text or images, it is more difficult to collect large amounts of high-quality, high-dimensional video data, let along text-video pairs.
Diffusion Models for Video Generation Said 12 Apr 2024
High-quality data collection depends more on careful human execution than on machine learning techniques alone.
Lots of ML techniques in the post can help with data quality, but fundamentally human data collection involves attention to details and careful execution.
Thinking about High-Quality Human Data Said 5 Feb 2024
Higher-perplexity adversarial attack text is easier to detect and mitigate.
High perplexity makes an attack more vulnerable to be detected and mitigated.
Adversarial Attacks on LLMs Said 25 Oct 2023
Universal adversarial trigger attacks are easy to detect because the learned trigger tokens tend to be nonsensical.
One drawback with UAT (Universal Adversarial Trigger) attacks is that it is easy to detect them because the learned triggers are often nonsensical.
Adversarial Attacks on LLMs Said 25 Oct 2023
What is a korrent?
A korrent is a belief a person has stated in their own words: one sentence stating the claim, backed by a quote and a source, kept at korrents.com.
Under a name here, the quoted block is what they actually said. The korrent beneath it is the claim those words support, in korrents' wording — tap it to see the record, its source, and who else holds it.
Nobody here wrote their own korrents. They are compiled from public statements, and a person can change their mind, which is recorded too.
Where they publish
Blog Lil'Log Long survey posts on how a model technique works. Has a feed.
A handful of posts a year, each a worked survey of one area — agents, hallucination, diffusion, reward hacking — with the maths kept in and the papers cited.
Recent
- Harness Engineering for Self-Improvement 4 Jul 2026 The concept of recursive self-improvement (RSI) dates back to I. J. Good (1965), where he defined an “ultraintelligent machine” as a system that can surpass humans in all intellectual activities and design b…
- Scaling Laws, Carefully 24 Jun 2026 Scaling laws are one of the most critical empirical findings in deep learning. The observation is simple in form: the training loss $L$ decreases predictably as we scale up model size $N$, dataset size $D$, and compute…
- Why We Think 1 May 2025 Special thanks to John Schulman for a lot of super valuable feedback and direct edits on this post. Test time compute (Graves et al. 2016, Ling, et al. 2017, Cobbe et al. 2021) and Chain-of-thought (CoT) (Wei et al. 202…
Show 17 more
- Reward Hacking in Reinforcement Learning 28 Nov 2024 Reward hacking occurs when a reinforcement learning (RL) agent exploits flaws or ambiguities in the reward function to achieve high rewards, without genuinely learning or completing the intended task. Reward hacking exi…
- Extrinsic Hallucinations in LLMs 7 Jul 2024 Hallucination in large language models usually refers to the model generating unfaithful, fabricated, inconsistent, or nonsensical content. As a term, hallucination has been somewhat generalized to cases when the model…
- Diffusion Models for Video Generation 12 Apr 2024 Diffusion models have demonstrated strong results on image synthesis in past years. Now the research community has started working on a harder task—using it for video generation. The task itself is a superset of t…
- Thinking about High-Quality Human Data 5 Feb 2024 [Special thank you to Ian Kivlichan for many useful pointers (E.g. the 100+ year old Nature paper “Vox populi”) and nice feedback. 🙏 ] High-quality data is the fuel for modern data deep learning model train…
- Adversarial Attacks on LLMs 25 Oct 2023 The use of large language models in the real world has strongly accelerated by the launch of ChatGPT. We (including my team at OpenAI, shoutout to them) have invested a lot of effort to build default safe behavior into…
- LLM Powered Autonomous Agents 23 Jun 2023 Building agents with LLM (large language model) as its core controller is a cool concept. Several proof-of-concepts demos, such as AutoGPT, GPT-Engineer and BabyAGI, serve as inspiring examples. The potentiality of LLM…
- Prompt Engineering 15 Mar 2023 , also known as In-Context Prompting, refers to methods for how to communicate with LLM to steer its behavior for desired outcomes without updating the model weights. It is an empirical science and the effect of prompt…
- The Transformer Family Version 2.0 27 Jan 2023 Many new Transformer architecture improvements have been proposed since my last post on “The Transformer Family” about three years ago. Here I did a big refactoring and enrichment of that 2020 post — r…
- Large Transformer Model Inference Optimization 10 Jan 2023 [Updated on 2023-01-24: add a small section on Distillation.] Large transformer models are mainstream nowadays, creating SoTA results for a variety of tasks. They are powerful but very expensive to train and use. The ex…
- Some Math behind Neural Tangent Kernel 8 Sept 2022 Neural networks are well known to be over-parameterized and can often easily fit data with near-zero training loss with decent generalization performance on test dataset. Although all these parameters are initialized at…
- Generalized Visual Language Models 9 Jun 2022 Processing images to generate text, such as image captioning and visual question-answering, has been studied for years. Traditionally such systems rely on an object detection network as a vision encoder to capture visua…
- Learning with not Enough Data Part 3: Data Generation 15 Apr 2022 Here comes the Part 3 on learning with not enough data (Previous: Part 1 and Part 2). Let’s consider two approaches for generating synthetic data for training. Augmented data. Given a set of existing training samples, w…
- Learning with not Enough Data Part 2: Active Learning 20 Feb 2022 This is part 2 of what to do when facing a limited amount of labeled data for supervised learning tasks. This time we will get some amount of human labeling work involved, but within a budget limit, and therefore we nee…
- Learning with not Enough Data Part 1: Semi-Supervised Learning 5 Dec 2021 When facing a limited amount of labeled data for supervised learning tasks, four approaches are commonly discussed.
- How to Train Really Large Models on Many GPUs? 24 Sept 2021 [Updated on 2022-03-13: add expert choice routing.] [Updated on 2022-06-10]: Greg and I wrote a shorted and upgraded version of this post, published on OpenAI Blog: “Techniques for Training Large Neural Networks&r…
- What are Diffusion Models? 11 Jul 2021 [Updated on 2021-09-19: Highly recommend this blog post on score-based generative modeling by Yang Song (author of several key papers in the references)]. [Updated on 2022-08-27: Added classifier-free guidance, GLIDE, u…
- Contrastive Representation Learning 31 May 2021 The goal of contrastive representation learning is to learn such an embedding space in which similar sample pairs stay close to each other while dissimilar ones are far apart. Contrastive learning can be applied to both…
Link verified 20 Sept 2026. Recent items update automatically from the channel.
x.com @lilianweng Posts.
Recent
- It is a hard and sad decision. I shared this message with folks at Thinky. Thank you all for the time together♥️ Just as the last sentence in my message: The future worth building is human. 27 Jul 2026
- Inkling is our open weights model. It aims to serve as a foundation with solid performance across a broad categories of capabilities, for use in practice and customization. Play it on Tinker! 😄 15 Jul 2026
- new post on harness engineering for AI self-improvement: https://t.co/ZYvGfVs61k It is hard to forecast how much the future of RSI will rely on harnesses. Likely harness engineering will evolve in the direction of self-improvement and enable auto-research, and, in turn, smarter models keeps harnesses simple. Even when many harness improvement get eventually internalized into core model, the need to specify goals and context will not disappear. 7 Jul 2026
Show 11 more
- A super long overdue (3+ years?) post on scaling laws. Compute is expensive. Scaling laws are a way to help us reason about the optimal compute allocation between data and model size before committing to a large run. The post covers what scaling laws predict, how compute-optimal allocation works, why Kaplan et al. and Chinchilla disagree, and how data limits + fitting details make extrapolation tricky. 25 Jun 2026 lilianweng.github.io
- We would love to see more collaboration and research in the field of human-AI interactivity. Check it out! 19 May 2026
- I only recently read more about the concept of system accidents by Charles Perrow, very insightful and relatable. 18 May 2026
- In the past few months, we had a lot of fun (and stress 😅) to produce 12 versions (+ many subversions) and 137 pages in our training run log book. Turns out human-human collaboration is important to improving human-AI collaboration. 😊 11 May 2026
- Building technologies for better human-AI collaboration on next gen hardware at scale. Exciting. 10 Mar 2026
- I’ve been telling people this a lot today: I enjoy so much working with people who care about what they are building and craftsmanship. It is a privilege to have a chance to work on something I’m passionate about, beyond making a living. I cherish it and don’t take it for granted. 15 Jan 2026
- On-policy distillation provides an elegant way to use the teacher model as a process reward model to provide dense reward while preventing SFT style "OOD shock" during rollout. 27 Oct 2025
- GPUs are expensive and setting up the infrastructure to make GPUs work for you properly is complex, making experimentation on cutting-edge models challenging for researchers and ML practitioners. Providing high quality research tooling is one of the most effective ways to improve research productivity of the wider community and Tinker API is one step towards our mission there. Tinker API is built on top of our experimental results on fine-tuning with LoRA: https://t.co/bV00V1cBKG Beta starts and you can join the waitlist today: 1 Oct 2025 thinkingmachines.ai
- Looking through those little hidden gem stories in the footnote, you will find it so inspiring that researchers with interests on the same topic are able to work together to advance a field despite their roles and locations. This is the power of open science and community. 26 Sept 2025
- Besides the fun fact that Connectionism is connected with the early days of the AI field and highlights similarities between neural networks and human brains, the flagship product of the (first) Thinking Machines is named Connection Machine. — 🧑🎓Enjoy reading and more is coming! 10 Sept 2025
- Welcome @_kevinlu to the team! Exciting to work together again. 18 Aug 2025
Link verified 20 Sept 2026. Recent items update automatically from the channel.
GitHub @lilianweng Code, and the source of the blog.
Recent
- lilianweng.github.io — My personal page 14 Jul 2026
- multi-armed-bandit — Play with the solutions to the multi-armed-bandit problem. 21 May 2024
- lil-log — Lilian's Blog 1 Sept 2023
Show 10 more
- transformer-tensorflow — Implementation of Transformer Model in Tensorflow 25 Mar 2023
- deep-reinforcement-learning-gym — Deep reinforcement learning model implementation in Tensorflow + OpenAI gym 24 Mar 2023
- emoji-semantic-search — Search the most relevant emojis given a natural language query 1 Jan 2023
- stock-rnn — Predict stock market prices using RNN model with multilayer LSTM cells + optional multi-stock embeddings. 28 Jul 2022
- generalization-experiment — TBA 22 Mar 2019
- unified-gan-tensorflow — A Tensorflow implementation of GAN, WGAN and WGAN with gradient penalty. 27 Aug 2018
- LeetcodePython — Leetcode Practice in Python 14 Oct 2014
- ScholarFame — Small analysis about fame and productivity of scholars 14 Feb 2014
- BagSaleDetector — Handbag Sale Detector 3 Aug 2012
- BlankTabMultivariateTestAddon — The merged add-on of new tab page with multiple designs, used for firefox testpilot study 2 Aug 2011
Link verified 20 Sept 2026. Recent items update automatically from the channel.
About the English under a post
Some people here publish in a language other than English. Where they do, this site shows a machine translation beneath the post, in this typeface — the site's own, not theirs.
The post itself is never changed, moved or hidden: what is set in the serif above is exactly what the person published, and it is what to quote them on. A translation can be wrong in ways that matter, especially about tone.
Only the post's own words are translated. A quoted post, a linked article and a belief on korrents.com are left in their original language.