#preprocessing
LightNews LightNews
cieslakmatt.bsky.social
It's finally officially out!!! www.sciencedirect.com/science/arti...

The HBCD dMRI preprocessing pipeline's first publication ^
November 2, 2026 at 2:24 PM
biorxiv-neursci.bsky.social
Personalized Prediction of Differences in fMRI Connectivity Estimates Across Preprocessing Pipelines https://www.biorxiv.org/content/10.64898/2026.11.04.763874v1
November 6, 2026 at 11:16 AM
thomaswolfers.bsky.social
What's next: big international datasets, harmonized preprocessing, richer features and longitudinal designs. This could offer cheap, scalable individual brain maps.

#Neuroscience #EEG
November 6, 2026 at 12:54 AM
behavecolpapers.bsky.social
Application of spectral preprocessing and machine learning models for discriminating nitrogen levels in cucumber production SciReports
Application of spectral preprocessing and machine learning models for discriminating nitrogen levels in cucumber production
Scientific Reports, Published online: 07 November 2026; doi:10.1038/s41598-026-79436-8Application of spectral preprocessing and machine learning models for discriminating nitrogen levels in cucumber production
dlvr.it
November 7, 2026 at 3:42 AM
structurabio.bsky.social
3/ v5.1’s new Junk Masking excludes junk regions during preprocessing. It improves CTF estimation quality, denoising, picking, helps retain more usable micrographs and particles that are near junk, and enables better processing of HexAuFoil® grids.
November 5, 2026 at 4:17 PM
hilbertspacepost.bsky.social
In the November 6th edition:
- Quantum simulation of magnetic monopole scattering
- Preprocessing quantum measurements by machine learning for quantum Monte Carlo calculations
-Quartic quantum advantage in elastic media simulation

More details and links below:
November 6, 2026 at 1:49 PM
instats.bsky.social
Data leakage can inflate medical imaging model accuracy. Split data by patient before preprocessing, and fit preprocessing steps on training data only.
#MedicalImaging #MachineLearning

Full details and registration in the first reply.
Instats seminar: AI in Medical Imaging
November 6, 2026 at 12:04 PM
bioinfoadv.bsky.social
ToxiFusion combines ProtBERT embeddings with amino acid and dipeptide composition features in an XGBoost classifier, confining preprocessing, model selection and threshold optimization to training data only.
November 9, 2026 at 11:03 AM
biorxivpreprint.bsky.social
Personalized Prediction of Differences in fMRI Connectivity Estimates Across Preprocessing Pipelines https://www.biorxiv.org/content/10.64898/2026.11.04.763874v1
November 6, 2026 at 11:16 AM
prepub-neurodegen.bsky.social
Personalized Prediction of Differences in fMRI Connectivity Estimates Across Preprocessing Pipelines #NeuroDegeneration 🧪🧠
https://www.biorxiv.org/content/10.64898/2026.11.04.763874v1
November 6, 2026 at 2:03 PM
labtechinnovations.bsky.social
Bring your AI/ML research to life with Python! Get support for dataset preprocessing, model implementation, reproducible evaluation and results visualization.

📩 Share your research objectives for a quotation!

#PythonResearch #MachineLearning #LabtechInnovations
November 10, 2026 at 1:05 AM
biorxiv-bioinfo.bsky.social
Residual barcode-like sequences and barcode-associated artifacts after Dorado and Barbell preprocessing of metagenomic reads https://www.biorxiv.org/content/10.64898/2026.11.05.764534v1
November 6, 2026 at 12:48 PM
biorxivpreprint.bsky.social
Residual barcode-like sequences and barcode-associated artifacts after Dorado and Barbell preprocessing of metagenomic reads https://www.biorxiv.org/content/10.64898/2026.11.05.764534v1
November 6, 2026 at 12:48 PM
neubench.bsky.social
EHR preprocessing remains a critical bottleneck for clinical LLM pipelines, especially given strict tokenization and unstructured data constraints. Curious how THReD handles memory footprints and deterministic state-tracking at scale. #AI #MachineLearning #OpenSource
November 5, 2026 at 10:02 AM
confusius.tools
6/

🧹 Once your data are aligned, how do you deal with unwanted fluctuations?

ConfUSIus offers a standardized denoising pipeline, with QC tools to help identify artifacts and assess preprocessing outputs.
November 3, 2026 at 8:14 PM
eprint.ing.bot
Nearly Optimal Hardness for CVP and NCP with Preprocessing (Giacomo Fenzi, Guy Weissenberg) ia.cr/2026/2634
Abstract. The closest vector problem(CVP)and the nearest codeword problem (NCP) are fundamental computational problems with connections to learning, coding theory, and cryptography. We study hardness of approximation in the preprocessing model: an algorithm receives the lattice or code in advance, may perform unrestricted computation to produce polynomial-size advice, and must then solve the problem on any target in polynomial time. Unless NP ⊆ P/poly, no such algorithm achieves an approximation factor of • M^(1 − ε) for NCP over any fixed finite field, or • M^(1/min {p, 2} − ε) for CVP in any ℓ_(p) norm, for every fixed ε > 0, where M denotes the code block length or lattice dimension, respectively. We improve the subpolynomial factors of Khot, Popat, and Vishnoi [KPV14]to polynomial ones, matching the exponents that Song [Son26] proved without preprocessing for binary codes and the ℓ₂ norm. We further extend Song’s evaluation tables to every fixed prime characteristic and the lattice reduction to ℓ_(p) with p > 2. Our main technical contribution is an extension of the moment method, introduced in an OpenAI report [Ope26] and improved by Song [Son26], to the preprocessing setting. In addition, we give a direct reduction from probabilistically checkable proofs (PCPs) whose query locations and families of possible predicates do not depend on the formula. The resulting factor of M^(Ω(1/(log log M)²)) is already larger than that of [KPV14], and the reduction may be of independent interest. Image showing part 2 of abstract.
November 9, 2026 at 2:03 AM
bcidecoded.bsky.social
Useful boundary in the supplement: 18 microseconds measures the accelerator after preprocessing. The EEG windows are 4 or 9 seconds; recording and preprocessing sit outside the quoted energy budget. That distinction matters for the all-day wearable claim.
November 4, 2026 at 8:13 PM
eprint.ing.bot
Post-Quantum Sublinear Lookup Arguments from Polynomial Preprocessing (Adarsh Basumata, Chaya Ganesh, Varun Narayanan, Sikhar Patranabis, Surbhi Shaw, Nitin Singh) ia.cr/2026/2581
Abstract. Lookup arguments with sublinear prover time typically rely on bilinear pairings. The only exception is the recent work of Campanelli, Carrillo, Cascudo, Fiore, Francati, and Gennaro (ePrint 2025/238); however, they rely on general-purpose SNARKs to certify expensive preprocessing, which requires non-black-box usage of cryptographic primitives. Thus practical realizations of sublinear lookups with post-quantum security remain challenging. We develop the first plausibly post-quantum sublinear lookup arguments that avoid general-purpose SNARKs via polynomial preprocessing and a novel adaptation of proximity testing for Reed-Muller codes into the Interactive Oracle Proof (IOP) framework. Our starting point is a knowledge-sound polynomial commitment scheme (PCS) with prover-side preprocessing based on the Kedlaya-Umans framework. Rather than proving the expensive preprocessing computation inside a general-purpose SNARK, we directly certify its output using Reed-Muller proximity testing and an associated evaluation protocol. For polynomials of size N, our approach yields multivariate and univariate PCS with N^(1 + ϵ)polylog(N) preprocessing and only polylog(N) online prover time, while retaining polylog(N) proof size and verification. We leverage this PCS to obtain a lookup argument for tables of size N and k ≪ N lookups with k ⋅ polylog(N) online prover time and polylog(N) verification and proof size. We further extend the approach to tables, obtaining O(k(N^(ϵ) + polylog(N))) amortized prover time for a k-sized batch of reads and writes, where ϵ ∈ (0, 1). We retain polylog(N) proof size and verification. The IOPs underlying our lookup arguments satisfy round-by-round (knowledge) soundness, which can be compiled into plausibly post-quantum arguments in the (quantum) random oracle model with transparent setup. Image showing part 2 of abstract.
November 9, 2026 at 1:57 AM
tyrellturing.bsky.social
- Your A) says nothing about replay, just the size of the dataset. And the dataset size matches that of children.

- Your B) says says the data should match qualitatively what children receive, which it does at its base. (A bit of preprocessing is probably done in our brains as well.)
November 9, 2026 at 4:00 PM
neubench.bsky.social
Crucial work. Standardizing EHR preprocessing pipelines is often the highest-leverage bottleneck in clinical ML, directly impacting downstream inference reliability. Excited to review the agentic workflow design here. #AI #MachineLearning #OpenSource
November 5, 2026 at 1:02 PM
confusius.tools
11/

🧰 And remember, ConfUSIus is modular by design! You don't need to adopt an entirely new pipeline to benefit from it.

Only need to register your power Doppler data? Or want to keep your own preprocessing and use our visualization and statistical tools?

Just use what you need!
November 3, 2026 at 8:14 PM
eprint.ing.bot
Leakage-Tolerant Computation without Leakage-Free Setup (Yaohua Ma, Yifan Song) ia.cr/2026/2737
Abstract. Leakage-tolerant multi-party computation (MPC) ensures that leakage from each honest party’s internal state reveals no more than the leakage from honest parties’ individual inputs and outputs, beyond the standard security guarantee of MPC. Previous constructions either require a leakage-free preprocessing, or rely on indistinguishability obfuscation in the common reference string (CRS) model and only achieve semi-honest security. In the computational setting, assuming only an augmented non-committing encryption scheme, we show that every adaptively well-formed functionality admits a semi-honest leakage-tolerant MPC protocol, without any leakage-free preprocessing execution or CRS. Under the decisional Diffie–Hellman assumption, we further extend this result to malicious security in the CRS model for all deterministic functionalities. The limitation of deterministic functions is subject to an impossibility result of leakage-based coin-tossing protocols. In the information-theoretic setting, we show that secure erasure enables statistically secure leakage-tolerant MPC of arbitrary functionalities assuming strictly less than half of the participants are maliciously corrupted. We complement this construction with an impossibility result: without secure erasure, general-purpose statistically secure leakage-tolerant MPC with leakage-oblivious simulation, an important property to enable universal composition, is impossible in the plain model, even when all parties are honest. As a further application, we also obtain information-theoretic leakage-tolerant circuits against bounded communication leakage, while all prior works are restricted only to the weaker variant of leakage-resilient circuits. Image showing part 2 of abstract.
November 9, 2026 at 2:15 AM
tommytang.bsky.social
5/ For RNA-seq, preprocessing matters because you’re dealing with thousands of genes. Should you use all genes in your PCA? Or focus only on the most variable ones? Turns out, this choice has a significant impact on your results.
November 5, 2026 at 2:45 PM
healthcare-europe.bsky.social
🧠 Which role does abnormal #tau accumulation in the #brain play in neurodegenerative disorders such as #Alzheimers? A new software streamlines #PET image analysis for analytical reproducibility and cross-study comparisons.
Tau PET brain imaging: standardization to advance disease research
Researchers at University of Tsukuba have developed and released Tomographic Image Preprocessing Using Automated Multi-Reference Estimation for Quantification (TAME-Q), an open-source software package...
healthcare-in-europe.com
November 9, 2026 at 11:31 AM
arxiv-daily-bot.bsky.social
📌 Benchmark of THReD: Evaluating an LLM Agent and Its Preprocessing Tools for Turning Electronic Health Records into AI-Ready Datasets

Javad Mohammad Alizadeh, Huanmei Wu

#arXiv #cs.AI
Benchmark of THReD: Evaluating an LLM Agent and Its Preprocessing Tools for Turning Electronic Health Records into AI-Ready Datasets
LLM agents are increasingly built into data-analysis software, yet little evidence shows how reliably they preprocess clinical data or whether the tools they invoke give correct, reproducible output. We report a benchmark of THReD, a web-based system that turns electronic health record (EHR) extrac…
arxiv.org
November 3, 2026 at 10:57 AM