I like these suggestions also for ARR and NLP conferences! Keep the rebuttal and score updates, remove the discussion, penalize reviewers who don’t update their scores.

Victor Escorcia@escorciav.bsky.social · 22h ago

Ok & kudos to @neuripsconf.bsky.social PCs 💪🏼 Diversity of trials is the best way to test what works & not. I doubt that any AI venue can claim to have nailed peer review at scale. Otherwise social media during review release & decision period would be 🌈💐

may be time to resurface this <squints> ten year old FATML paper that explored the ethics of automatic experimentation, say by agents. 841.io/doc/ethics-o...

Alexandra Lahav@alahav.bsky.social · 2h ago

New @bloomberglaw.com article by Olivia Carville - TikTok experimented on users, putting 10% (about 15M people) in a "filter bubble" exposing them to repetitive content that the algorithm thought they wanted, but was toxic. If you want to know why these cases are going to be massive...

To whoever wrote this, thank you for saying it and know that it helps a bit. And I'm still so sorry and frustrated that we have silenced international students.

E.g., our reasoning work here: academic.oup.com/pnasnexus/ar... shows how (even much older) models and humans make similar patterns of trading off between logic and semantic biases that can roughly be interpreted as "using prior knowledge. 3/

Language models, like humans, show content effects on reasoning tasks

Abstract. Abstract reasoning is a key ability for an intelligent system. Large language models (LMs) achieve above-chance performance on abstract reasoning

academic.oup.com

as we get further into the end of days, I'm reminded about how much of research taste, values and judgment is bound up in the taste, values and judgment of small communities of researchers

I've been so busy that I missed the tech report release!! this model is pretty interesting. it is FAST!! +1000 tokens per second! (I saw +3K in a demo) and changing from Auto regressive to Diffusion is pretty cool!

Sung Kim@sungkim.bsky.social · yesterday

Google DeepMind's DiffusionGemma Technical Report They feel text diffusion models open up a radically different part of the latency–quality Pareto frontier and hope the report makes it easier for researchers and engineers to understand the model, build on it, and create things we haven’t thought of

📣 I am hiring a 𝗣𝗼𝘀𝘁𝗱𝗼𝗰 𝗶𝗻 𝗦𝗼𝗰𝗶𝗮𝗹 𝗥𝗲𝗶𝗻𝗳𝗼𝗿𝗰𝗲𝗺𝗲𝗻𝘁 𝗟𝗲𝗮𝗿𝗻𝗶𝗻𝗴 & 𝗛𝘂𝗺𝗮𝗻-𝗔𝗜 𝗛𝘆𝗯𝗿𝗶𝗱 𝗦𝘆𝘀𝘁𝗲𝗺𝘀 📍 Berlin 📅 3-year position 💡 Ideal for researchers with a PhD in Computer Science, Cognitive Science, Applied Math, Physics, etc. and a strong background in AI, RL, and computational social science. Details below

Postdoctoral Fellow in Social Reinforcement Learning and Human-AI Hybrid Systems - Max-Planck-Institut für Bildungsforschung

The Center for Humans and Machines (CHM) at the Max Planck Institute for Human Development in Berlin conducts interdisciplinary science to understand, anticipate, and shape major disruptions from A...

careers.mpib-berlin.mpg.de

There are two wolves. One always lies. One controls the lever of trolley track. They're in a state of superposition inside a box. On the trolley are infinite monkeys at infinite typewriters going to an infinite hotel. The trolley must first travel halfway, but before that a quarterway, and so on...

Left, right, whatever, I just want graduate students posting their papers here

Yes, the result itself is very interesting (more on this below). From a skim, I agree with Noah that the paper is not well-written. Am I happy to see LLMs invade TCS? No, it's terrible. Is this the most impactful LLM result about lattices this week? Unclear. 1/

Clément Canonne@ccanonne.github.io · 4d ago

(same thing for @huckbennett.bsky.social and @noahsd.bsky.social , regarding the CVP one... If you have the time and energy of course!)

We are delaying general registration to Aug 10 to carefully manage author registration and make sure that all authors (main + workshops) get to register. We appreciate the patience

If you’re curious about why people are so excited/concerned about the state of agents, this is a really nice tutorial. If you think this stuff is useless and you’re open to changing your mind, this might be interesting!

David P. Reichert@david-p-reichert.bsky.social · 2d ago

Many don't see how much AI has changed recently. Coding agents aren't just chatbots. We can’t deal with the challenges of AI if we don’t understand it. Here's a post to provide some evidence (Claude doing small ML experiments) and form intuitions. davidpreichert.substack.com/p/if-you-hav... 🧵..

So… they’re working

If your paper was desk rejected from ACL 2026 at the Camera Ready stage due to hallucinated references, please read this important note from the #EMNLP2026 Program Committee. 🔗 Details: 2026.emnlp.org/concerning-l...

"We're all worried," as what it means to do research (in my field, Theoretical CS) seems to be shifting, and shifting fast. What to do? Senior researchers must lead by example, knowing that not everything will pan out. What I'm suggesting below may not work everywhere, but here's my own advice: 1/

Some initial thoughts, and a complicated mix of feelings. Wow. I mean, Erdos problems are cool (I genuinely mean that), I didn't know about the Jacobian conjecture before it got disproved. But this newest batch from OpenAI hits home in a way the previous announcements did not.

I wrote about science funding, science reform, and anti-science agendas. scatter.wordpress.com/2026/07/31/s...

science funding and the cherry picking problem

Science publishing–like all of science–requires money. There’s no getting around it, even with the open science movement. What appears free to many authors and readers still requires money and infr…

scatter.wordpress.com

The deadline has been extended to Monday, August 3 --

Denis@dpeskoff.bsky.social · 3w ago

Do you work across computational methods, social sciences, and the humanities? Submit to Text as Data 2026! 📄 One-page submissions 🔓 Non-archival ⏰ Due August 1 📍 October 5 @UCBerkeley tada2026.org

“…[COLM] may have to narrow the scope of allowed contributions. Theoryslop and slopterpretability are hard to manage in the same review process as other papers.” I recognize these categories, but the idea of punishing all theory and framework papers is tough. Agreed on more desk rejections.

Conference on Language Modeling@colmweb.org · 5d ago

📢 New blog post describing the COLM 2026 PCs’ analyses of AI use in submitted papers. gregdurrett.github.io/colm2026-blo... (temporary home, new COLM website coming soon!)

📢 New blog post describing the COLM 2026 PCs’ analyses of AI use in submitted papers. gregdurrett.github.io/colm2026-blo... (temporary home, new COLM website coming soon!)

New Blog Post: "The Fate of the Hobbyist Programmer" stephenramsay.net/posts/fate-o...

If you're going to cite that NBER report from OpenAI about "how people use AI," you've got to at least caveat it. We have no way to directly verify most of the report, and we do have at least one good reason, via indirect evidence and quoted below, to not trust it. www.nber.org/papers/w34255

Maria Antoniak@mariaa.bsky.social · 7mo ago

Not the point of the OP, but I was curious and took a closer look at this 2025 piece from OpenAI. Much could be said, but it fails my go-to test for all of these pieces about "how people actually use AI": the tokens "sex" "erotic" and "NSFW" do not appear in the paper.

One weird trick to literally destroy America www.wsj.com/politics/pol...

Exclusive | U.S. Weighs $100,000 Fee for Foreign Students Wanting to Work After Graduation

The proposed policy would particularly affect Silicon Valley, Wall Street and universities.

wsj.com

@katestarbird.bsky.social is my hero. @ic2s2.bsky.social

Me when "book guillotines" hits the discourse again

ICLR 2027 has authorship quotas: - No more than 20 submissions per author - No more than 1 submission where no author has a paper previously accepted to a major ML conference From iclr.cc/Conferences/...

Wonderful talk by @sjgreenwood.bsky.social about @paper-feed.bsky.social! Custom feeds on Bluesky are unique resources for researchers interested in recommendation algorithms #IC2S2 Featuring many Bluesky friends like @graze.social @devingaffney.com @kissane.myatproto.social @aendra.com

“…our sense that we are reading the work of another human being, someone with connections to others, to a time and a place like postwar Naples, someone to whom we might relate our own life stories, is our strongest reason not to cede too much ground to AI-generated text.”

Dan Cohen@dancohen.org · 6d ago

New issue of my newsletter Humane Ingenuity: “The Breath of the Author” — The palpable presence of someone else’s mind in our best writing should give us pause about the encroachment of AI text newsletter.dancohen.org/archive/the-...

PSA: VADER was created over a decade ago, and we have many newer and better tools for sentiment analysis, including efficient methods (mostly fine-tuned BERT variants). It's not hard to tune your own model, and there are 15k (!) models ready to go on Hugging Face! huggingface.co/models?searc...

🚨 Call for #EMNLP2026 Student Volunteers Volunteer for 10 hours and receive FREE conference registration Both in-person and virtual roles are available 📅 Deadline: September 1 📢 Decisions: September 9 🔗 Link: docs.google.com/forms/d/e/1F... More info: 2026.emnlp.org/calls/volunt...