@AdenIshaq Ishaq Aden-Ali

Ishaq Aden-Ali posts on X about this is, llm, preference, target the most. They currently have [---] followers and [--] posts still getting attention that total [-----] engagements in the last [--] hours.

Engagements: [-----] #

Mentions: [--] #

Followers: [---] #

CreatorRank: [---------] #

Social Influence

Social category influence celebrities

Social topic influence this is, llm, preference, target, hidden, in the, ai, citizen, breaking, what are

Top accounts mentioned or mentioned by @nhaghtal @axliu42 @golowichnoah @ashettyv @jeffdean @minilek @fermima @vaishnavh @yangpliu @pfau @chrispainteryup @alzzyd @elonmusk @druv_pai

Top Social Posts

Top posts by engagements in the last [--] hours

"@pfau @ChrisPainterYup What are some of the main technical and social (political) road blocks"
X Link 2025-11-03T19:55Z [---] followers, [--] engagements

"@alz_zyd_ Neals take is definitely more correct than yours"
X Link 2025-11-23T02:27Z [---] followers, [---] engagements

"RT @JeffDean: This is completely not okay and we can't become numb to repeated instances of illegal and unconstitutional action by governm"
X Link 2026-01-10T20:40Z [---] followers, [---] engagements

"1/7 Excited about our new paper with @axliu42 @GolowichNoah @AShettyV @nhaghtal and Ankur on how data selection can have wild effects"
X Link 2026-02-05T23:17Z [---] followers, 36.1K engagements

"6/7 The correlations are all very small/near-orthogonal but due to the linearity they add up and fine-tuning on the preference dataset causes the model to behave as if it were system-prompted This can get quite interesting"
X Link 2026-02-05T23:17Z [---] followers, [---] engagements

"7/7 Here's another fun example: we can get the fine-tuned model to really love Elephants and mention them frequently in every context. Check out our paper for more examples and all the details"
X Link 2026-02-05T23:17Z [---] followers, [---] engagements

"RT @axliu42: First post on X excited to share this new work"
X Link 2026-02-05T23:19Z [---] followers, [--] engagements

"RT @nhaghtal: Did you know that your LLM is secretly an Ouija board Fun fact: Subsets of standard data sets can embed hidden instructions"
X Link 2026-02-06T01:45Z [---] followers, [--] engagements

"RT @minilek: "As a concrete example of our technique. there is a simple way to choose a subset of a standard preference learning dataset"
X Link 2026-02-06T03:10Z [---] followers, [--] engagements

"RT @fermi_ma: This is a very neat result: given a dataset and a target system prompt like reply in Spanish they show you can select a su"
X Link 2026-02-06T22:34Z [---] followers, [--] engagements

"RT @nhaghtal: 1/n Now that I have a bit more time I wanted to share more about this paper and my own thought process. Theres a gap in h"
X Link 2026-02-06T23:09Z [---] followers, [--] engagements

"RT @_vaishnavh: enjoyed how this paper ties up two prior observations that are equally mindboggling (low-rank logits & subliminal/weird gen"
X Link 2026-02-13T20:47Z [---] followers, [--] engagements

"RT @yangpliu: 1/ Technical thread on #1stProof Problem 6: finding spectrally light vertex subsets in a graph and how its solution fits i"
X Link 2026-02-15T03:34Z [---] followers, [--] engagements

"1/ Technical thread on #1stProof Problem 6: finding spectrally light vertex subsets in a graph and how its solution fits into the landscape of spectral sparsification + restricted invertibility. Original thread: https://x.com/yangpliu/status/2022690162220716327 My thoughts on #1stProof Problem [--] (closely related to areas I've worked in): OpenAIs solution is essentially correct and the difficulty feels consistent with AI capabilities over the past several months. More detail in the thread. https://x.com/yangpliu/status/2022690162220716327 My thoughts on #1stProof Problem [--] (closely related to"
X Link 2026-02-14T18:53Z [---] followers, 38K engagements

"My thoughts on #1stProof Problem [--] (closely related to areas I've worked in): OpenAIs solution is essentially correct and the difficulty feels consistent with AI capabilities over the past several months. More detail in the thread"
X Link 2026-02-14T15:11Z [---] followers, 69.3K engagements

"enjoyed how this paper ties up two prior observations that are equally mindboggling (low-rank logits & subliminal/weird generalization effects). 1/7 Excited about our new paper with @axliu42 @GolowichNoah @AShettyV @nhaghtal and Ankur on how data selection can have wild effects https://t.co/r2fYujg1GX 1/7 Excited about our new paper with @axliu42 @GolowichNoah @AShettyV @nhaghtal and Ankur on how data selection can have wild effects https://t.co/r2fYujg1GX"
X Link 2026-02-13T20:27Z [----] followers, [----] engagements

"1/7 Excited about our new paper with @axliu42 @GolowichNoah @AShettyV @nhaghtal and Ankur on how data selection can have wild effects"
X Link 2026-02-05T23:17Z [---] followers, 36.1K engagements

"RT @nhaghtal: 1/n Now that I have a bit more time I wanted to share more about this paper and my own thought process. Theres a gap in h"
X Link 2026-02-06T23:09Z [---] followers, [--] engagements

"1/n Now that I have a bit more time I wanted to share more about this paper and my own thought process. Theres a gap in how we talk about data and behavior in LLMs. On the one hand we say data is the driver and try to interpret what human preferences the data is showing. On the other hand we keep seeing learned behaviors that dont seem to be present in the data at least not in any human-readable way. Examples like subliminal transfer covert malicious finetuning and weird generalization have been quite intriguing for me. In this paper we ask whats really causing that and whether theres a"
X Link 2026-02-06T21:04Z [----] followers, 10.7K engagements

"Did you know that your LLM is secretly an Ouija board Fun fact: Subsets of standard data sets can embed hidden instructions into your model and to turn them into evil rulers animal lovers and translators. No sys prompt. No signals. Just ghosts in the data. 1/7 Excited about our new paper with @axliu42 @GolowichNoah @AShettyV @nhaghtal and Ankur on how data selection can have wild effects https://t.co/r2fYujg1GX 1/7 Excited about our new paper with @axliu42 @GolowichNoah @AShettyV @nhaghtal and Ankur on how data selection can have wild effects https://t.co/r2fYujg1GX"
X Link 2026-02-06T00:45Z [----] followers, 34.9K engagements

"RT @fermi_ma: This is a very neat result: given a dataset and a target system prompt like reply in Spanish they show you can select a su"
X Link 2026-02-06T22:34Z [---] followers, [--] engagements

"This is a very neat result: given a dataset and a target system prompt like reply in Spanish they show you can select a subset of the data such that fine-tuning an LLM on that subset causes the model to behave as if it were given that system prompt 1/7 Excited about our new paper with @axliu42 @GolowichNoah @AShettyV @nhaghtal and Ankur on how data selection can have wild effects https://t.co/r2fYujg1GX 1/7 Excited about our new paper with @axliu42 @GolowichNoah @AShettyV @nhaghtal and Ankur on how data selection can have wild effects https://t.co/r2fYujg1GX"
X Link 2026-02-06T22:23Z [---] followers, [----] engagements

"RT @minilek: "As a concrete example of our technique. there is a simple way to choose a subset of a standard preference learning dataset"
X Link 2026-02-06T03:10Z [---] followers, [--] engagements

""As a concrete example of our technique. there is a simple way to choose a subset of a standard preference learning dataset. which contains no examples written in Spanish but when we fine-tune a model on this subset itl learns to speak primarily in Spanish." 🤯 Really cool 1/7 Excited about our new paper with @axliu42 @GolowichNoah @AShettyV @nhaghtal and Ankur on how data selection can have wild effects https://t.co/r2fYujg1GX 1/7 Excited about our new paper with @axliu42 @GolowichNoah @AShettyV @nhaghtal and Ankur on how data selection can have wild effects https://t.co/r2fYujg1GX"
X Link 2026-02-06T02:24Z 29.1K followers, [----] engagements

"RT @nhaghtal: Did you know that your LLM is secretly an Ouija board Fun fact: Subsets of standard data sets can embed hidden instructions"
X Link 2026-02-06T01:45Z [---] followers, [--] engagements

"RT @axliu42: First post on X excited to share this new work"
X Link 2026-02-05T23:19Z [---] followers, [--] engagements

"First post on X excited to share this new work 1/7 Excited about our new paper with @axliu42 @GolowichNoah @AShettyV @nhaghtal and Ankur on how data selection can have wild effects https://t.co/r2fYujg1GX 1/7 Excited about our new paper with @axliu42 @GolowichNoah @AShettyV @nhaghtal and Ankur on how data selection can have wild effects https://t.co/r2fYujg1GX"
X Link 2026-02-05T23:18Z [--] followers, [----] engagements

"This is completely not okay and we can't become numb to repeated instances of illegal and unconstitutional action by government agencies. The recent days have been horrific. 🚨BREAKING: Border Patrol just illegally arrested a U.S. citizen a TEENAGER in Richfield Minnesota. The teen was working at Target when agents tackled him and arrested him while his passport was in his pocket fully identifying himself as a U.S. citizen. None of it https://t.co/6xM6xEVHxD 🚨BREAKING: Border Patrol just illegally arrested a U.S. citizen a TEENAGER in Richfield Minnesota. The teen was working at Target when"
X Link 2026-01-09T19:45Z 409.8K followers, 1M engagements

"🚨BREAKING: Border Patrol just illegally arrested a U.S. citizen a TEENAGER in Richfield Minnesota. The teen was working at Target when agents tackled him and arrested him while his passport was in his pocket fully identifying himself as a U.S. citizen. None of it mattered. Why Because he looked Latino. When a coworker asked who to call he said heartbreakingly: my mom. Think about that. Your teenage son goes to work and Border Patrol kidnaps him. This isnt an isolated incident. ICE and Border Patrol have repeatedly illegally arrested U.S. citizens especially in Minnesota. And after agents"
X Link 2026-01-08T20:59Z 48.6K followers, 5.6M engagements

"this year's grad school angst post is about killing projects & what makes a research paper justifiable"
X Link 2025-12-28T17:45Z [----] followers, 20.1K engagements

"I swear whats happening now is the height of humiliation. Over [-----] tents are flooding at this very moment children the elderly and families all trapped inside. Bedding is soaked the cold is unbearable and rainwater has mixed with sewage washing away the last traces of dignity. We film we document we share it with the worldbut no one responds. Its as if the world has gotten used to these scenes as if our suffering has become part of their daily routine. Were not asking for the impossible just shelter to protect our lives. Simple caravans to shield us from the rain and cold. Nothing more. Yet"
X Link 2025-12-27T18:39Z 19.1K followers, 149.3K engagements

"Me rephrasing my unsolved math problem for the 37th time in an LLM: 🎰 This ones gotta hit"
X Link 2025-12-23T12:55Z [----] followers, 10.8K engagements

"Provably Learning from Modern Language Models via Low Logit Rank https://ift.tt/zphxI4w https://ift.tt/zphxI4w"
X Link 2025-12-11T05:04Z 24.6K followers, [----] engagements

Limited data mode. Full metrics available with subscription: lunarcrush.com/pricing