#interpretable
If science isnt interpretable, it wasn't done. Science is a narrative endeavor. It is human. It is political. Science is not data. It is the story about the data. It is the story about which data to get.
November 20, 2026 at 10:13 PM
Pollen Ornamentation Morphospace: Interpretable Quantification by Persistent Homology https://www.biorxiv.org/content/10.64898/2026.11.18.767721v1
November 20, 2026 at 4:48 AM
tomorrow's sunday! 🤓 i'm gonna take my photo tonight. just postin as a reminder for anyone else who might wanna keep an eye out for a neat shot for tomorrow.

("night lights" — interpretable however. 🤸🏻‍♂️)
#sundayshots "bird/s ~elsewhere~"

(next week: "night lights" 🤓)
November 21, 2026 at 8:17 PM
It's going to be funny when the aliens show up, hand us a manual for how the human mind works, and it turns out it's actually very easily interpretable but just requires way more steps than anyone had ever bothered trying
November 20, 2026 at 6:29 AM
AIO Part I: Interpretable ADMET Prediction for Cyclic Peptides and Macrocycles https://www.biorxiv.org/content/10.64898/2026.11.18.767182v1
November 19, 2026 at 8:47 PM
This was joint work with @scheon.com @shresearchucsd.bsky.social @mds71.bsky.social and my advisor @berkustun.bsky.social.

I'll present it at the Interpretable Deep Learning seminar on Wed, Nov 25, 8:30am PT (www.youtube.com/@Interpretab...), and at #NeurIPS2026 in Sydney (Tue 10am, poster S512).
Interpretable Deep Learning
Interpretable models expose the internal reasoning of a system in a form that is understandable to humans. Once limited by accuracy, they now achieve performance comparable to powerful opaque models s...
www.youtube.com
November 19, 2026 at 11:18 PM
Show your support to the trans community by posting a transgender character ! 🏳️‍⚧️
Like, basically all the Mirror Modulationlings are meant to be trans or easily interpretable as such but he's the most direct one about it. He's trans and it rocks!
November 18, 2026 at 5:25 PM
Accurate and interpretable protein complex stoichiometry prediction through agentic AI with StoiAgent https://www.biorxiv.org/content/10.64898/2026.11.17.767325v1
November 18, 2026 at 2:45 PM
Jaaa, ist auch in der @tagesschau.de.web.brid.gy angekommen: #Anthropic/C ist der größte Börsengang der Geschichte gelungen. Liegt vor #SpaceX.

Fand die Rede von Dario Amodei gut.
Bzgl. Chancen & Risiken.

www.anthropic.com
Home
Anthropic is an AI safety and research company that's working to build reliable, interpretable, and steerable AI systems.
www.anthropic.com
November 19, 2026 at 7:18 PM
Samitha up next, PhD student I co-supervise. Talking about an interpretable multi-omcis deeplearning framewoek for gene dependency modelling in Paediatric Brain Cancer (Poster #30)
@tyagilab.bsky.social

#ABACBS
#ABACBS26
#ABACBS2026
November 17, 2026 at 11:19 PM
Dawg hasn’t had an interpretable radio message since like Azerbaijan
November 20, 2026 at 4:10 AM
Pollen Ornamentation Morphospace: Interpretable Quantification by Persistent Homology https://www.biorxiv.org/content/10.64898/2026.11.18.767721v1
November 21, 2026 at 2:04 AM
funny though that the original black boxes were quite interpretable and it was a thought experiment about “what if they weren’t”
November 18, 2026 at 1:59 AM
Pollen Ornamentation Morphospace: Interpretable Quantification by Persistent Homology https://www.biorxiv.org/content/10.64898/2026.11.18.767721v1
November 19, 2026 at 8:01 PM
Pollen Ornamentation Morphospace: Interpretable Quantification by Persistent Homology https://www.biorxiv.org/content/10.64898/2026.11.18.767721v1
November 19, 2026 at 8:01 PM
Read the Cancer Research article "Counterfactual Diffusion Models Provide Interpretable Explanations of Artificial Intelligence Models in Pathology," by ESMO AI & Digital Oncology Annual Congress Scientific Cochair Jakob Kather: buff.ly/SrWhwMt @jnkt.bsky.social #OpenAccess #ESMOAI26
November 17, 2026 at 8:01 AM
⚪️ my queer childhood memories are not vague and interpretable, they're shit like "mom, i want to be a lesbian when i grow up"
November 15, 2026 at 5:05 AM
A Toolkit For Automated And Interpretable Characterization Of Material Properties From Spectral Data!
November 19, 2026 at 7:26 AM
AIO Part I: Interpretable ADMET Prediction for Cyclic Peptides and Macrocycles https://www.biorxiv.org/content/10.64898/2026.11.18.767182v1
November 19, 2026 at 8:47 PM
Oh my! Yeah, just seeing that news.

Sounds hugely exciting. Hugely messy.

Also, I kind of get the pressure to push a proof out. Wish I had the confidence that they'd continue to invest in producing something interpretable that the community can benefit from.
November 20, 2026 at 7:46 PM
And Second:
'PSTR235.15 / FF9 - Interpretable decoding of reach kinematics from ECoG using a CNN-Transformer-LSTM model'
November 15, 2026 at 5:37 PM
2026 has been incredibly kind to our research group when it comes to publications! 🎉

Happy to share a new paper where we use interpretable machine learning to predict nitrate & phosphorus and identify the watershed conditions contributing to nutrient pollution. 🌎💧

🔗 doi.org/10.1016/j.je...
November 20, 2026 at 4:56 PM
RLDE is a reinforcement learning framework for interpretable iterative document editing. It formulates document revision as a reference-free MDP problem. Evaluations showed statistically significant results across all metrics.

Source: arXiv cs.CL
RLDE: A Reinforcement Learning Framework for Interpretable Iterative Document Editing
Abstract page for arXiv paper 2611.22039: RLDE: A Reinforcement Learning Framework for Interpretable Iterative Document Editing
arxiv.org
November 18, 2026 at 10:26 AM
/ es que no es comparable la documentación de una mitología a la otra, ni mucho menos la claridad de la misma :/ por eso es mucho más interpretable
November 17, 2026 at 1:51 PM