Markdown version: https://ai.hiina.space/reports/personal-web-surfing-models/09-sources.md ยท index
Index of everything on this site: https://ai.hiina.space/index.md
Personal
web surfing models and an ought-driven tab feed: 9. Sources
9. Sources
All pages below were read by the research passes, except where the
text says a figure came from a secondary source. Three items are dated
2026 and were read but not independently re-verified: Intent Assistant,
Compass and the Prosocial Ranking Challenge. Chang et al. (CHI 2021)
figures come from CMU's write-up and ScienceAlert because the paper
would not load.
- Knowledge: public per-person
link libraries
- Gwern, Guardian
Angels
- Screencasts
could be scalable data and evals for single-user emulation (LessWrong,
2026)
- Gwern,
virtual LessWrong comments (2024)
- Fu
& Pirolli, SNIF-ACT (2007)
- West
& Leskovec, human wayfinding in information networks (WWW
2012)
- Gellert,
web page prediction with Markov and PPM models (2021 preprint)
- GoWeB: next-page
prediction on Edge logs
- Google,
how machine learning improved Chrome (2024)
- Chrome
Topics API
- blink-dev:
Topics API removal thread
- GRU4Rec: session-based
recommendations with RNNs
- SASRec vs BERT4Rec
replicability study
- HSTU:
trillion-parameter sequential transducers (ICML 2024)
- TIGER:
generative retrieval for recommender systems (NeurIPS 2023)
- TALLRec
- LLaRA
- USER-LLM
(Google Research blog)
- ONCE: LLM-boosted
content-based recommendation
- MIND
news recommendation dataset
- EB-NeRD news
dataset
- WebShop
- Mind2Web
- WebArena
- WebLINX
(ICML 2024)
- Online-Mind2Web
- OPeRA: real-user shopping
browsing dataset (2025)
- Lu et al., simulating
real users' web shopping
- LaMP: personalized LLM
benchmark
- PRISM alignment
dataset
- Park et al., generative
agent simulations of 1,000 people
- PRISM-X (2026)
- PersonalWAB
- Persona2Web
(2026)
- Brank
et al., predicting per-user browsing trails (MSR 2004)
- Lettkeman
et al., predicting task-specific webpages for revisiting (AAAI
2006)
- Compass: realigning
short-video feeds to reflective preferences (2026)
- SANS,
Google Chrome forensics
- Chromium
visit annotations database source
- Firefox
Interactions.sys.mjs source
- Decoding
Firefox session store data
- snss crate: Chrome SNSS
session files
- chrome.tabs
extension API
- Promnesia
- ActivityWatch
web watcher add-on
- ArchiveBox
- Screenpipe
- Dwell time
normalization for implicit feedback
- Radlinski et al.,
natural-language user profiles for transparent recommendation
- Ramos et al.,
LLM-written user profiles (ACL 2024)
- LFM: short user profiles
for pairwise preference
- LUNAR: retrieval of
past records for personalization (2026)
- Hou et al., LLMs are
zero-shot rankers for recommender systems
- Ji et al., data leakage
in recommender evaluation
- Mozilla
Foundation, uniqueness and re-identifiability of browsing
histories
- IMDEA,
sensitive and tracked content across a billion URLs
- Anthropic
API data retention policy
- Anthropic,
updates to consumer terms
- Memex (Wikipedia;
Bush, "As We May Think")
- Knowledge
README
- Are.na API
- Hypothesis,
viewing and exporting annotations
- Memex
by WorldBrain add-on
- Histre, why
- RecAgent
- Agent4Rec (SIGIR
2024)
- USimAgent
- GUM: general user
models from screen activity (2025)
- TwinVoice
(2025)
- gwern.net build
code: GTX.hs
- gwern.net build
code: GenerateSimilar.hs
- Gwern newsletter,
November 2022
- gwern.net
repository
- CMU,
overcoming tab overload (on Chang et al., CHI 2021)
- ScienceAlert,
tab overload survey
- Dubroy
& Balakrishnan, tabbed browsing study (CHI 2010)
- Tabs.do (UIST
2021)
- Arc,
archived tabs help article
- BuzzFeed,
when people actually read later (2013)
- Obendorf
et al., web page revisitation revisited (CHI 2007)
- 9to5Mac,
Mozilla announces Pocket shutdown
- Kleinberg, Mullainathan
& Raghavan, the challenge of understanding what users want
- Milli et al.,
engagement, user satisfaction and divisive content on Twitter
- Milli et al., choosing
the right weights for behavioral signals
- Jia, Bernstein et al.,
embedding societal values into social media feeds (CSCW 2024)
- Prosocial Ranking
Challenge results (2026)
- BONSAI:
intent-specified Bluesky feeds (2025)
- Readwise Reader,
August 2026 update
- Choi et al., Intent
Assistant (CHI 2026)
- Haliburton
et al., one sec friction study (CHI 2024)
- Lukoff et al., designing
for user agency on YouTube (CHI 2021)
Text licensed CC BY 4.0 (attribution: Hiina, ai.hiina.space). Source:
https://github.com/hiinaspace/ai-hiina-space