Professor, musician, speaker of @neuroexplicit. This account is inactive - please find me at bsky.app/profile/akoller.bsk…
Joined March 2009
- Tweets 191
- Following 239
- Followers 1,215
- Likes 332
4 Photos and videos
15 Apr 2025
This is a great opportunity to work with world-class faculty. The 16 PhD students who have already joined are excellent and fun. I hope to hear from you! #NLP #nlproc
Final opportunity: Multiple PhD students to start in Fall 2025. Combine neural and symbolic/interpretable models of language, vision, and action, and work with world-class advisors at @Saar_Uni, MPI Informatics, @mpi_sws, @CISPA ,@DFKI. Details: neuroexplicit.org/jobs/
1
3
490
26 Feb 2025
Now hiring: Multiple PhD students to start in fall 2025, for research on combining neural and symbolic/interpretable models of language, vision, and action. Work with world-class advisors at @Saar_Uni, MPI Informatics, @mpi_sws, @CISPA ,@DFKI. Details: neuroexplicit.org/jobs/
3
5
943
10 Jan 2025
I am following my university in leaving Twitter. I would be very pleased if you chose to reconnect with me at bsky.app/profile/akoller.bsk…
See you there!
researchprofessionalnews.com…
1
219
16 Dec 2024
AutoPlanBench 2.0 now evaluates LLMs as planners on more than 50 domains. ReAct (with GPT-4o) is often worse, but sometimes better than symbolic planners. coli-saar.github.io/autoplan… #nlproc
1
3
362
11 Oct 2024
Come do research with me, my fantastic colleagues, and some of the coolest PhD students I've ever met! #NLProc #nesy #neurosymbolic #AI #ML
Now hiring: Twelve (!) PhD students to start in fall 2025, for research on combining neural and symbolic/interpretable models of language, vision, and action. Work with world-class advisors at @Saar_Uni, MPI Informatics, @mpi_sws, @CISPA, @DFKI. Details: neuroexplicit.org/jobs
1
2
10
2,692
Alexander Koller retweeted
10 Oct 2024
spending a great week in Saarland for the @neuroexplicit retreat of @SIC_Saar.
great talks, posters and interactions with #PhD students on #neurosymbolic #nesy #AI #ML
(and boardgames!)
Thanks @alkoller @slusallek @IValeraM for having me!
3
19
2,052
19 Aug 2024
So proud of my postdoc Mareike Hartmann for this excellent work.
14 Aug 2024
🏆 ACL Best Resource Paper Award:
AppWorld: A Controllable World of Apps and People for Benchmarking Interactive Coding Agents by Trivedi et al.
#NLProc #ACL2024NLP
1
1
17
941
12 Jul 2024
It was fun to apply #NLProc methods to software engineering with my brilliant colleague @AndreasZeller and his student @TurikMammadov. The coolest part, to me, is that you can backtranslate program outputs into program inputs. Let's see where this will go!
12 Jul 2024
Learning models from programs! Given a program P, our MODELIZER learns a model M that mocks P's behavior, producing P's output for a given input. But M is also reversible, predicting inputs for which P produces a given output, with up to 95.4% accuracy: arxiv.org/abs/2407.08597 🧵
ALT An input generator synthesizes inputs for the program under test, which then produces outputs. From the input/output pairs, a Model Learner creates a model, which can then translate unseen inputs into unseen outputs and vice versa.
1
6
890
1 Jul 2024
Can you use LLMs to replace crowdworkers in NLP evaluations? My amazing collaborators and I analyzed this broadly. Answer: Sometimes LLMs correlate very well with human judgments, but you can't rely on it.
1 Jul 2024
1/5 📣 Excited to share “LLMs instead of Human Judges? A Large Scale Empirical Study across 20 NLP Evaluation Tasks”! arxiv.org/abs/2406.18403 🚀 We introduce JUDGE-BENCH, a benchmark to investigate to what extent LLM-generated judgements align with human evaluations. #NLProc
1
5
30
3,547
14 Jun 2024
Ellie is fantastic, and I am so delighted to have her here.
Ellie Pavlick @Brown_NLP is visiting us for three months as a @dfg_public Mercator Fellow. We are thrilled to have her with us and look forward to many fruitful research collaborations.
3
465
Alexander Koller retweeted
2 Jun 2024
This is a decent summary of the octopus thought experiment from Bender & @alkoller 2020, with two glaring exceptions, right at the start:
techcrunch.com/2024/06/01/wt…
>>
2
20
82
16,626
3 Jun 2024
Come work with my amazing colleagues and me on neurosymbolic models for #NLProc and related fields! You'll join our really excellent first nine PhD students and one of the largest research centers for neurosymbolic models in the world. @LstSaar @SIC_Saar
Now hiring: Three PhD students and one postdoc, as part of a research team of ~30, combining neural and symbolic/interpretable models of language, vision, action, and ML. Work with advisors at @Saar_Uni, MPI Informatics, @mpi_sws_, @CISPA, @DFKI. Details: neuroexplicit.org/jobs/
7
12
5,980
15 Dec 2023
ChatGPT getting information from
Bild - what could possibly go wrong?
We have formed a new global partnership with @AxelSpringer and its news products.
Real-time information from @politico, @BusinessInsider, European properties @BILD and @welt, and other publications will soon be available to ChatGPT users.
ChatGPT’s answers to user queries will include attribution and links to full articles for transparency and further information. openai.com/blog/axel-springe…
5
939
15 Dec 2023
Come work with my amazing colleagues and me on neurosymbolic models! You'll join our really excellent first six PhD students and one of the largest research centers for neurosymbolic models in the world. #NLProc @LstSaar @SIC_Saar
Now hiring: Six PhD students and one postdoc, for research on combining neural and symbolic/interpretable models of language, vision, and action and ML. Work with world-class advisors at @Saar_Uni, MPI Informatics, @mpi_sws_, @CISPA, @DFKI. Details: neuroexplicit.org/jobs/
1
6
888
16 Nov 2023
Can LLMs do planning? My PhD student @Stein1Katharina built AutoPlanBench, which can automatically convert any PDDL benchmark domain into a benchmark for LLM planners, and they are not doing so hot. coli-saar.github.io/autoplan… #NLProc
1
1
7
909
16 Nov 2023
My student @yuekun_yao did something really cool: Predict accuracy of a seq2seq model on test data from only the inputs. Core is a discriminator that learns to check whether the model's prediction is correct. Excellent accuracy across datasets. coli-saar.github.io/discrimi… #NLProc
1
2
11
2,242
15 Nov 2023
A very fun piece of work that I got to collaborate on at @allen_ai: Hierarchical plans improve LLM planning on domains that have hierarchical structure.
13 Nov 2023
How can LLM-agents dynamically adapt to task complexity & LLM capabilities "as-needed"?
🚨Introducing ADaPT to recursively intervene/decompose if task is too complex -> yields substantial gains on interactive tasks
arxiv.org/abs/2311.05772
w/ @ai2_aristo @allen_ai @uncnlp
🧵⬇️
1
1
8
2,109
Alexander Koller retweeted
4 Oct 2023
Nucleus and top-k sampling are ubiquitous, but why do they work?
@johnhewtt, @alkoller, @swabhz, @Ashish_S_AI and I explain the theory and give a new method to address model errors at their source (the softmax bottleneck)!
📄 arxiv.org/abs/2310.01693
🧑💻 github.com/mattf1n/basis-awa…
3
26
161
32,782
3 Aug 2023
Come work with me: Three-year postdoc position, very suitable for developing your own research agenda and collaborations. Let's figure out reliable reasoning with LLMs and personalization of text and dialogue. Neurosymbolic models welcome. coli.uni-saarland.de/~koller…
18
42
6,920
14 Jul 2023
Come work with us on neurosymbolic models of #NLProc! The first three students are amazing - be part of a wonderful team that investigates the design principles of combining neural and symbolic models. #ACL2023 #ACL2023NLP
Now hiring: Three more PhD students and one postdoc for research on combining neural and symbolic/interpretable models of language, vision, and action. Work with world-class advisors at
@Saar_Uni, MPI Informatics, @mpi_sws,
@CISPA, @DFKI. Details: neuroexplicit.org/jobs
4
11
2,129