Summarize Paper logo SummarizePaper.com
  • Rising topics
  • Weekly letters
  • Read papers faster

    Summaries Chat with papers Collections

    Follow your field

    Daily paper alert Scoop watch Premium Citation alerts Premium Find a research placement Premium+ Who works on it? Premium Rising topics Weekly letters Saved searches

    Write and submit

    Press kit Premium+ Literature review Premium+ What did I miss? Premium Bibliography check Premium Zotero, Overleaf, Notion, Obsidian

    Present and meet

    Journal club Premium+ Conference companion Premium+
    All the tools, with what they do
  • FAQ
  • Premium
  • Français English
  • Login

Rising topics in research

Every week, the topics that appear in many more new papers than usual on arXiv, bioRxiv and medRxiv, from several independent teams, explained in plain language.

Is my topic rising? How it works Illustrated guide

arXivbioRxivmedRxiv For you Premium

Mathematics (math): 4 rising topics this week, with Premium

  • Predicting how wireless signals travel
  • Tracking waves with mathematical weights
  • Understanding physics through conformal symmetry
  • How compressible fluids settle and respond to control

Below, the free topics of the whole server.

Discover Premium · US$9 / month
All of arXiv Computer Science (cs) Condensed Matter (cond-mat) Electrical Engineering and Systems Science (eess) Mathematical Physics (math-ph) Mathematics (math)

All of arXiv

Papers submitted from 2 November to 8 November 2026

← Previous week

What the papers found, field by field: the weekly letters.

Sudden rise

Faster ways to analyze changing networks

×3.3 compared with usual · 85 papers from 68 teams

7.7 per 10,000 papers7.5 per 10,000 papers5.3 per 10,000 papers3.8 per 10,000 papers4.1 per 10,000 papers4.2 per 10,000 papers7.5 per 10,000 papers4.3 per 10,000 papers5.6 per 10,000 papers7.6 per 10,000 papers6.1 per 10,000 papers6.3 per 10,000 papers5.3 per 10,000 papers6.7 per 10,000 papers8.6 per 10,000 papers10.4 per 10,000 papers4.0 per 10,000 papers8.8 per 10,000 papers7.5 per 10,000 papers6.9 per 10,000 papers8.5 per 10,000 papers3.4 per 10,000 papers7.9 per 10,000 papers11.5 per 10,000 papers7.9 per 10,000 papers13.6 per 10,000 papers27.4 per 10,000 papers

New this week: often an event, a release, a data set or a deadline. The coming weeks will tell if it lasts.

Researchers are developing algorithms to find routes, identify bottlenecks, and connect selected points in networks, including networks whose links change over time. The increased activity spans related questions: how quickly answers can be updated, how much information must be stored, and how much approximation is acceptable.

undirected graphundirectedstretchupdate timevertex graph
  • Finding and Certifying Weighted Minimum Cuts with Cut Queries
  • Dynamic Steiner Tree under Edge Updates
  • First Near-Additive All-Pairs Approximate Shortest Paths in Decremental Weighted Graphs
  • Faster Dynamic Maximal Independent Set against Adaptive Adversaries
85 papers Ask these papers Who works on it? Literature review Share
More
  • Search these papers and follow the term: save the search to get an email when new papers use it.
  • Scoop watch: if your own project is close, be told the day a new paper comes close to it.
  • Find a research placement: who leads this work, for an internship or a PhD.
  • The history of its terms: compared week by week, to export.
Share “Faster ways to analyze changing networks” A post ready for LinkedIn, X or Bluesky
LinkedIn X Bluesky Email

X and Bluesky get a shorter post, unless you edit it above. The link shows a preview picture of this page.

Comeback

Changing AI behavior through its internal weights

×3.8 compared with usual · 19 papers from 19 teams

6.4 per 10,000 papers12.6 per 10,000 papers4.0 per 10,000 papers5.1 per 10,000 papers5.4 per 10,000 papers2.8 per 10,000 papers3.0 per 10,000 papers4.3 per 10,000 papers7.0 per 10,000 papers0.0 per 10,000 papers0.0 per 10,000 papers4.7 per 10,000 papers4.0 per 10,000 papers10.7 per 10,000 papers5.8 per 10,000 papers5.9 per 10,000 papers5.3 per 10,000 papers0.0 per 10,000 papers3.8 per 10,000 papers5.8 per 10,000 papers5.3 per 10,000 papers10.3 per 10,000 papers6.9 per 10,000 papers3.1 per 10,000 papers6.8 per 10,000 papers0.0 per 10,000 papers20.8 per 10,000 papers

Already this frequent the week of 11 May 2026.

A model’s weights are the numerical settings learned during training; weight-space research studies how changing these settings changes its behavior. The examples suggest growing interest in more targeted adaptation and auditing, from improving image restoration to understanding what useful information fine-tuning adds.

weight-space
  • Weight-Space Programming for Improving Image-Restoration Diffusion Models: Technical Conjectures and a Recursive Improvement Loop
  • Weight-space programming for auditing clinical decision networks
  • The Behavioral Rate of Fine-Tuning: Compression as a Measure of Useful Information
  • Can an LLM Actually Reason and Program about Neural Network's Weights or is it just memorization on steroids? Exploring "Weight-Space Programming"
19 papers Ask these papers Who works on it? Literature review Share
More
  • Search these papers and follow the term: save the search to get an email when new papers use it.
  • Scoop watch: if your own project is close, be told the day a new paper comes close to it.
  • Find a research placement: who leads this work, for an internship or a PhD.
  • The history of its terms: compared week by week, to export.
Share “Changing AI behavior through its internal weights” A post ready for LinkedIn, X or Bluesky
LinkedIn X Bluesky Email

X and Bluesky get a shorter post, unless you edit it above. The link shows a preview picture of this page.

Sudden rise

Making AI cheaper with lower-precision numbers

×2.7 compared with usual · 30 papers from 29 teams

2.6 per 10,000 papers7.5 per 10,000 papers9.3 per 10,000 papers12.7 per 10,000 papers2.7 per 10,000 papers5.6 per 10,000 papers6.0 per 10,000 papers5.7 per 10,000 papers5.6 per 10,000 papers4.6 per 10,000 papers7.7 per 10,000 papers4.7 per 10,000 papers10.6 per 10,000 papers4.0 per 10,000 papers4.3 per 10,000 papers7.4 per 10,000 papers6.6 per 10,000 papers12.6 per 10,000 papers2.5 per 10,000 papers8.1 per 10,000 papers13.8 per 10,000 papers12.8 per 10,000 papers13.9 per 10,000 papers15.7 per 10,000 papers10.2 per 10,000 papers16.7 per 10,000 papers32.9 per 10,000 papers

New this week: often an event, a release, a data set or a deadline. The coming weeks will tell if it lasts.

BF16 is a numerical format that reduces the memory needed to store and process neural-network values compared with standard 32-bit arithmetic. Its increased appearance fits ongoing efforts to make training and inference more efficient, while checking how numerical precision affects accuracy and practical abilities such as coding.

bf16
  • Beyond Accuracy: How Quantization Affects Coding Agents
  • Architecture-Aware Mixed Precision for Replica-Exchange Training of Geometry-Aware Neural Operators with Uncertainty Quantification
  • Dory1.0-4B: A Looped Normalized-GPT
  • ResQAdam: Residual-Preserving Adaptive Optimization for Memory-Efficient Quantization Aware Training
30 papers Ask these papers Who works on it? Literature review Share
More
  • Search these papers and follow the term: save the search to get an email when new papers use it.
  • Scoop watch: if your own project is close, be told the day a new paper comes close to it.
  • Find a research placement: who leads this work, for an internship or a PhD.
  • The history of its terms: compared week by week, to export.
Share “Making AI cheaper with lower-precision numbers” A post ready for LinkedIn, X or Bluesky
LinkedIn X Bluesky Email

X and Bluesky get a shorter post, unless you edit it above. The link shows a preview picture of this page.

Share this week's rising topics on arXiv A post ready for LinkedIn, X or Bluesky
LinkedIn X Bluesky Email

X and Bluesky get a shorter post, unless you edit it above. The link shows a preview picture of this page.

What Premium adds on this page

  • The rising topics of each field — 21 fields of arXiv have their own this week, down to categories such as cs.LG or astro-ph.EP.
  • For you — the topics closest to your research first, in the fields of your paper alert.
  • In your inbox — the rising topics of your fields in your Monday paper alert.
  • History of terms — three terms compared week by week, in any field, with the counts to download.
Discover Premium · US$9 / month
Next edition — papers submitted from 9 November to 15 November 2026.
Calculation starts no earlier than Tuesday 17 November 2026 at 07:00 Paris time (06:00 UTC). Results appear once processing is complete.

How it works

We count arXiv, bioRxiv and medRxiv separately: each server, and each of their fields, has its own rising topics. Journal articles are not counted.

Each week, we count how many papers submitted that week (first versions) use each term of one to three words in their title and abstract; acronyms such as RAG or JWST count, and plurals with their singular. A term rises when its share of papers is at least twice its share over the eight weeks before, when so large a rise is very unlikely by chance (a log-likelihood ratio of at least 15), and when it comes from at least three teams, none signing more than half of the papers: one group posting a series is not a trend. An AI then groups the rising terms into topics and names them; it never decides what rises: the counts do.

  • New: the term was almost absent from these papers before.
  • Rising for several weeks: up three weeks in a row or more, not a one-week event.
  • Recurring: as frequent at this time last year: perhaps a yearly cycle, such as a conference deadline.
  • Comeback: as frequent some months ago, then less used.
  • Sudden rise: new this week: often an event, a release, a data set or a deadline.

The curve on each topic is the share of papers using its main term, per 10,000 papers, over the last six months. Counting the previous Monday-to-Sunday submission week starts on Tuesday at 06:00 UTC, after the normal Monday arXiv announcement. Delayed announcements may still be missing from the first count.

© 2026 SummarizePaper · Open source on GitHub

  • Rising topics
  • Weekly letters
  • Guides
  • Examples
  • About
  • For institutions
  • Contact
  • Privacy Policy
  • Legal Notice
  • Text and data mining