Index of everything on this site: https://ai.hiina.space/index.md
No one has published a model trained on one person's open-web history and used to rank links or tabs for them or an agent. Several neighbouring fields are mature: next-page prediction, session-based recommendation, web-agent benchmarks and personal LLMs. Each covers part of what a personal surfing model needs.
| Field | Representative work | What it gives | Why it isn't a personal surfing model |
|---|---|---|---|
| Cognitive models of link choice | SNIF-ACT (Fu & Pirolli 2007): picks links by "information scent" relative to a goal. West & Leskovec (WWW 2012): Wikispeedia players jump to a hub, then steer by similarity | An interpretable account of why a goal-directed person clicks a link. Usable as features or a baseline | Population-level and goal-given; no personal taste |
| Next-page prediction | Markov and PPM models (Gellert 2021); GoWeB on Edge logs; Chrome omnibox ML scoring (Google, 2024) | Proof that history predicts the next destination | Predicts habit, which is the opposite of "ought" |
| History to interest profile | Chrome Topics API: hostnames to 22 root topics, top 5 per week. Being removed in 2026 | A production design for compressing history | Deliberately coarse, built for ads |
| Sequential recommendation | GRU4Rec, SASRec, BERT4Rec (replicability study), HSTU (Meta, 1.5T parameters in production), TIGER | Architectures for next-item prediction | Need a shared catalogue and millions of users. One person's web has an open item space and thousands of events |
| LLM recommenders | TALLRec (works with under 100 labels), LLaRA, USER-LLM (compresses long history to 32 soft tokens), ONCE (GPT infers a profile from news history) | The building blocks for an n=1 ranker | Trained and evaluated on pooled users |
| News click data with dwell | MIND, EB-NeRD (dwell and scroll depth) | Public data for pretraining a ranker or testing label noise | News only, other people |
| Web navigation agents | WebShop, Mind2Web, WebArena, WebLINX, Online-Mind2Web | The research agent a personal ranker would steer | Optimize for finishing tasks, not for taste |
| Imitating real browsing people | OPeRA (2025: 51 real users, about 4 weeks of Amazon logs with in-the-moment rationales and personas); Lu et al. (3,526 users) | The closest data and evaluation recipe | Exact next action is hard: 12 to 22% accuracy. Rationales help. No per-user model |
| Personal LLMs and digital twins | LaMP, PRISM; Park et al. (1,052 interview-based agents reach 85% of people's own test-retest consistency); PRISM-X (2026: per-user tuning gains small) | Evidence that individual imitation works on attitudes and is hard on preferences | Built from interviews, chat or writing, not browsing |
| Personalized web agents | PersonalWAB (purchase histories); Persona2Web (2026, synthetic year-long browsing histories) | A benchmark shape for "agent conditioned on my history" | Persona2Web agents reach at most about 13%. Inferring preferences from implicit history is the hard part, and explicit profiles help most |
Two things are worth knowing beyond the table:
The only n=1 browsing models I found are from the 2000s. Brank et al. (MSR 2004, 30 users) predicted per-user habitual trails. Lettkeman et al. (AAAI 2006) predicted per-task revisits. Both were small, and association rules failed on individual logs.
Compass (August 2026) is the nearest feed-side work. An LLM realigns a YouTube Shorts feed to the user's reflective preferences, with prompts during browsing. A 10-day study with 15 users reported more purposeful use.
The gap has three parts. First, a model of one person's long-term open-web history evaluated as a ranker for an agent. Second, a split between "would click" and "endorse" for a personal tab queue. Third, any shareable per-person model. I tried the wordings personal, single-user, n-of-1, tab recommendation, revisitation and read-later. This looks like a real gap, not a search miss.
Text licensed CC BY 4.0 (attribution: Hiina, ai.hiina.space). Source: https://github.com/hiinaspace/ai-hiina-space