160,000+ papers scored by impact · Conferences & journals ranked · Research gaps surfaced · Find where to submit your next paper.
1,000 top papers per section · updated daily
Each surface leans on the same scoring engine — recency, novelty, citation quality, venue rank, author reputation, completeness.
Papers scored by impact, novelty, and venue rank. Filter by topic, year, difficulty, or source.
A* venues (NeurIPS, ICML, ACL…) and Q1 journals with papers, deadlines, and rankings.
Paste title + abstract — get ranked conference or journal matches with deadlines and reviewer expectations.
80+ research topics clustered by theme with reading paths from Beginner to Frontier and trend scores.
Gaps extracted from papers — explicit, pattern-based, and starter ideas. Find your next direction.
Researchers ranked by momentum — who is publishing the most impactful work right now.
Paste title + abstract and get ranked conference matches with deadlines, acceptance rates, and reviewer expectations — plus live abstract-structure analysis and a sample draft to try.
Match your manuscript against 20 Q1 journal profiles — impact factor, review timeline, open access, and similar published papers, with a Copy-shortlist export.
We index 60,000+ papers from arXiv CoRR and 20+ major conference proceedings — all served from a Railway PostgreSQL backend so you can browse and search the full dataset with no caps.
Pull every new CS paper from arXiv CoRR (19 categories) plus proceedings from NeurIPS, ICML, ICLR, CVPR, ACL, KDD, AAAI and 15+ more venues.
Each paper gets a Paper Score 0–10 across 6 signals: recency, novelty, citation quality, institution prestige, author reputation, and completeness.
60,000+ papers land in Railway PostgreSQL — fully browsable and searchable with server-side filtering and pagination, no caps.
arXiv papers update daily (Mon–Fri). Conference papers are permanent — added once per proceedings cycle, never expire.
A paper published this year scores 10/10. One from 10+ years ago scores 0. CS moves fast — freshness matters.
We scan the title and abstract for words like "we propose", "outperforms", "state-of-the-art". Surveys and replications score lower.
Log-scaled citation count plus conference rank (A* = top-tier), with an extra boost for oral/spotlight acceptances. A NeurIPS oral with 500 citations scores near 10.
Papers from OpenAI, DeepMind, Google Research, MIT, Stanford, CMU etc. get a boost. Detected from affiliations or abstract text.
Papers co-authored by Turing Award winners, lab founders, or well-known researchers (Hinton, LeCun, Vaswani, Karpathy…) score higher.
Has the paper got an abstract, author list, PDF link, and topic tags? Well-documented papers surface above poorly indexed ones.
Author momentum measures how productive and impactful a researcher has been recently — not just their all-time record. It combines recent output (30%), average paper quality (25%), acceleration vs. their prior two-year window (20%), topic strength in trending areas (15%), and conference strength (10%). A senior professor with 10,000 citations but no papers in 3 years scores low. A PhD student publishing 4 strong LLM papers this year at NeurIPS scores high. Momentum finds who is actively pushing the frontier right now.
Three-theme interface — Atelier Zero, Industrial Brutalist & Field Notes
Choose your reading environment from the nav — the warm editorial Atelier Zero, a high-contrast Industrial Brutalist, or the tactile Field Notes. Your pick persists across every page and reload with no flash of the wrong theme. Ships with a rebuilt responsive navigation (grouped menus, search, auth controls, mobile) and a modular CSS architecture for shared foundations, components, motion, and contrast.
OpenReview acceptance tiers — Oral & Spotlight signals
Papers from ICLR, NeurIPS, ICML, and COLM now carry their acceptance tier (oral / spotlight / poster). Oral and spotlight papers get a score boost and show as badges. Coverage extended through ICLR 2026, NeurIPS 2025, and ICML 2025.
Smarter Conference & Journal recommenders
Both recommenders now run on one shared matching engine. New: one-click sample abstracts, field auto-detection with a confidence level, a fit-scale legend, ⌘/Ctrl+Enter to run, auto-saved drafts, and a Copy-shortlist export.
Keyless journal fetching via OpenAlex
The pipeline can now pull every paper from 20 top CS journals (TPAMI, IJCV, AIJ, PR, TKDE, Nature Communications…) directly through OpenAlex by source ID — no Semantic Scholar API key required. S2 is still used automatically when a key is configured.
Visual refresh — Atelier Zero skin
Site-wide rebrand: warm paper canvas, Inter Tight + Playfair Display typography, single coral accent, italic-serif Roman numerals, side rails and metadata strip, hairline borders. Now the default of the three-theme interface.
Dedicated Sign-in & Register pages
Sign-in and Create Account are now separate focused pages instead of a shared tabbed modal. Each page redirects back to where you came from after auth.
Security hardening — Auth & API
JWT secret misconfiguration now raises a loud server warning; API errors return a generic message instead of leaking internal details; SSL mode is conditional for local dev; login no longer saves partial auth state on profile-fetch failure.
Journal recommender + 20 Q1 journals
New Journal Recommender matches your manuscript against Q1 journal profiles — impact factor, review timeline, open access status, and similar published papers. Covers JMLR, TMLR, TACL, TPAMI, Nature MI, CSUR, and 14 more.
Backend API on Railway + user accounts
FastAPI backend deployed on Railway with PostgreSQL full-text search. Sign up to save favourites and sync across devices. API available at railway.app/docs.
OpenAlex integration + HF training dataset
Papers now sourced from OpenAlex (250M+ works) in addition to arXiv, S2, ACL, OpenReview, PMLR, and CVF. Full dataset pushed to HuggingFace as JSONL for LLM training.
Complete A* conference coverage
AAAI, IJCAI, CHI, SIGIR, WWW, KDD, WSDM, SIGMOD, ICSE now bulk-fetched via Semantic Scholar alongside ICLR, NeurIPS, ICML, CVPR, ACL, EMNLP and more.
CiteLens integration — citation analysis in one click
Every arXiv paper and Paper of the Day has an "Analyze citations" button linking to CiteLens — see who cited it, ranked by impact and relevance.