Jacob Morrison
@jacobcares
Followers
487
Following
25K
Media
15
Statuses
184
PhD student @uwnlp, research @allen_ai @ai2_allennlp
Seattle
Joined April 2009
I'm so excited that we're finally releasing Tülu 3, our new post-training recipe! We're releasing models built on top of Llama 3.1 base (OLMo coming soon!), all of our datasets, a (73 page!) paper, new evaluations, and all of our code.
Meet Tülu 3 -- a set of state-of-the-art instruct models with fully open data, eval code, and training algorithms. We invented new methods for fine-tuning language models with RL and built upon best practices in the community to scale synthetic instruction and preference data.
1
16
92
when you're passed out on the couch after eating turkey and loads of carbs and someone's trying to tell you it's time for pie:
45
4K
50K
🇺🇸🇺🇸🇺🇸🇺🇸🇺🇸
0
1
10
A number of people are talking about implications of AI to schools. I spoke about some of my thoughts to a school board earlier, some highlights: 1. You will never be able to detect the use of AI in homework. Full stop. All "detectors" of AI imo don't really work, can be
Gemini Nano Banana Pro can solve exam questions *in* the exam page image. With doodles, diagrams, all that. ChatGPT thinks these solutions are all correct except Se_2P_2 should be "diselenium diphosphide" and a spelling mistake (should be "thiocyanic acid" not "thoicyanic") :O
918
2K
16K
saumya is applying to PhD programs this year, hire her!!!!!!!!
Olmo 3 is out!!!! It was so much fun working on post-training. Loved seeing this come together with the best team!!!!
0
0
9
Olmo 3 is out!!!! It was so much fun working on post-training. Loved seeing this come together with the best team!!!!
Announcing Olmo 3, a leading fully open LM suite built for reasoning, chat, & tool use, and an open model flow—not just the final weights, but the entire training journey. Best fully open 32B reasoning model & best 32B base model. 🧵
1
5
75
it's finally here everybody, thanks for the patience
Announcing Olmo 3, a leading fully open LM suite built for reasoning, chat, & tool use, and an open model flow—not just the final weights, but the entire training journey. Best fully open 32B reasoning model & best 32B base model. 🧵
0
0
5
@MayeeChen @_awettig @heinemandavidj @scottgeng00 @VictoriaWGraf @jacobcares @saumyamalik44 (to my immeasurable sadness, @heinemandavidj and @saumyamalik44 are applying to PhDs this cycle; don’t be a fool, admit them!)
2
3
20
it's finally here!!!!
Announcing Olmo 3, a leading fully open LM suite built for reasoning, chat, & tool use, and an open model flow—not just the final weights, but the entire training journey. Best fully open 32B reasoning model & best 32B base model. 🧵
0
0
16
Announcing Olmo 3, a leading fully open LM suite built for reasoning, chat, & tool use, and an open model flow—not just the final weights, but the entire training journey. Best fully open 32B reasoning model & best 32B base model. 🧵
51
327
2K
Today, we are releasing the best open-weight LLM by a US company: Cogito v2.1 671B. On most industry benchmarks and our internal evals, the model performs competitively with frontier closed and open models, while being ahead of any US open model (such as the best versions of
5
32
649
🔥Thrilled to introduce DR Tulu-8B, an open long-form Deep Research model that matches OpenAI DR 💪Yes, just 8B! 🚀 The secret? We present Reinforcement Learning with Evolving Rubrics (RLER) for long-form non-verifiable DR tasks! Our rubrics: - co-evolve with the policy model -
7
116
537
No better time to learn about that #AI thing everyone's talking about... 📢 I'm recruiting PhD students in Computer Science or Information Science @Cornell_Bowers! If you're interested, apply to either department (yes, either program!) and list me as a potential advisor!
18
105
539
Forget modeling every belief and goal! What if we represented people as following simple scripts instead (i.e "cross the crosswalk")? Our new paper shows AI which models others’ minds as Python code 💻 can quickly and accurately predict human behavior! https://t.co/1t2fsW7jyL🧵
4
33
101
this is super cool 👀
👀 How to find more difficult/novel/salient evaluation data? ✨ Let the data generators find it for you! Introducing Data Swarms, multiple data generator LMs collaboratively search in the weight space to optimize quantitative desiderata of evaluation.
0
0
1
Evaluating language models is tricky, how do we know if our results are real, or due to random chance? We find an answer with two simple metrics: signal, a benchmark’s ability to separate models, and noise, a benchmark’s random variability between training steps 🧵
📢 New paper from Ai2: Signal & Noise asks a simple question—can language model benchmarks detect a true difference in model performance? 🧵
4
54
237
if you’re talking about zoo-related stuff and you start talking about how “zoos need to be abolished” i’m immediately not listening to you
215
5K
125K
America needs to take open models more seriously. This summer the early lead in open model adoption of the US via Llama has been overtaken by Chinese models. With The American Truly Open Models (ATOM) Project we're looking to build support and express the urgency of this issue.
35
104
631
The bottleneck in AI isn't just compute - it's access to diverse, high-quality data, much of which is locked away due to privacy, legal, or competitive concerns. What if there was a way to train better models collaboratively, without actually sharing your data? Introducing
Introducing FlexOlmo, a new paradigm for language model training that enables the co-development of AI through data collaboration. 🧵
8
47
255
Introducing FlexOlmo, a new paradigm for language model training that enables the co-development of AI through data collaboration. 🧵
13
75
436