CS Papers Deep-Read
Milestone Papers, Distilled — BigCat's Shelf
> one paper, one diagram · read this page ≈ understand the paper
Deep Learning · Modern AI
— neural nets / Transformers / LLMs / alignment
Paper 1
Attention Is All You Need — one attention-only architecture that became the foundation of every modern large model
Vaswani et al. · 2017
Paper 2
Deep Residual Learning (ResNet) — one skip connection let networks go hundreds of layers deep; the backbone of almost every deep net
He et al. · 2015
Paper 3
AlexNet — a deep net that learned its own features from a million images, won 2012 by a landslide and ignited the deep learning revolution
Krizhevsky et al. · 2012
Paper 4
Word2Vec — a machine that teaches itself word vectors from massive text; "king − man + woman ≈ queen"; where the embedding era began
Mikolov et al. · 2013
Paper 5
ViT (An Image is Worth 16×16 Words) — slice an image into patches as "words" for a standard Transformer; vision and language now share one architecture
Dosovitskiy et al. · 2020
Paper 6
CLIP — a matching game over 400M image–text pairs puts pictures and words in one shared meaning space; the foundation of multimodal AI
Radford et al. · 2021
Paper 7
LSTM (Long Short-Term Memory) — a memory conveyor plus two learned gates cured vanishing gradients and ruled sequence modeling for twenty years
Hochreiter & Schmidhuber · 1997
Paper 8
Seq2Seq (Sequence to Sequence Learning) — two LSTMs in relay, "understand then re-say," establishing the encoder-decoder paradigm
Sutskever, Vinyals & Le · 2014
Paper 9
Bahdanau Attention (Jointly Learning to Align and Translate) — let the decoder glance back at the source by relevance for every word it writes; attention's birth certificate
Bahdanau, Cho & Bengio · 2014
Paper 10
BERT — swap "guess the next word" for "fill in the blank" to read text both ways and teach itself language, founding the "pre-train + fine-tune" era
Devlin et al. · 2018
Paper 11
GPT-3 — scale "guess the next word" to 175B parameters and it picks up new tasks from a few examples; the start of the prompting era
Brown et al. · 2020
Paper 12
Scaling Laws — a model's skill falls along a smooth power law in "model + data + compute," turning big models from alchemy into engineering
Kaplan et al. · 2020
Paper 13
Emergent Abilities — some skills appear "suddenly" once a model crosses a scale threshold and can't be predicted; it also drew the famous "is it a mirage?" rebuttal
Wei et al. · 2022
Paper 14
InstructGPT / RLHF — align GPT-3 to human preference so it answers you thoughtfully, shifting the focus from "make it bigger" to "align it"; ChatGPT's recipe
Ouyang et al. · 2022
Paper 15
Constitutional AI — write the values into a short "constitution" and let the AI self-critique and rewrite against it (RLAIF), using almost no human harmfulness labels; a cornerstone of training Claude
Bai et al. · 2022
Paper 16
GAN (Generative Adversarial Networks) — a "forger" makes images and an "inspector" judges them; two adversaries push each other stronger, igniting a whole wave of generative modeling
Goodfellow et al. · 2014
Paper 17
DDPM (Denoising Diffusion Probabilistic Models) — sprinkle an image into static over a thousand steps, then train a network to spot the noise added at each one and run the chain backwards — where diffusion overtook GANs
Ho et al. · 2020
Batch Normalization — standardize each layer's inputs with the current mini-batch's own mean and variance, and training deep networks stops being tightrope walking
Ioffe & Szegedy · 2015
Dropout — send half the neurons home at random on every training step, so none can lean on a partner and one weight set buys a whole ensemble
Srivastava et al. · 2014
Adam (Adaptive Moment Estimation) — keep two ledgers per parameter, "which way on average" and "how big and jittery the slope is," to auto-scale each step; deep learning's default optimizer for a decade
Kingma & Ba · 2014
DQN (Deep Q-Network) — feed only screen pixels and score, use a convolutional network to estimate a "total future score" per action, and stabilize training with "experience replay"; one program learns 7 Atari games and founds deep reinforcement learning
Mnih et al. · 2013
AlphaGo — a policy network cuts away most bad moves and a value network skips reading to the end, guiding a tree search to first beat a professional in full-board Go
Silver et al. · 2016
Google Distributed Systems
— the original trio · the second-gen trio · evolution
Paper 1
The Google File System (GFS) — one master keeps the ledger, data sliced into big chunks stored three times; petabyte-scale storage from failure-prone cheap machines, big-data's first foundation stone
Ghemawat et al. · 2003
Paper 2
MapReduce — write just two functions, map + reduce, and the framework auto-parallelizes across thousands of machines with end-to-end fault tolerance, opening the big-data era
Dean & Ghemawat · 2004
Paper 3
Bigtable — a sparse sorted table keyed by (row, column, timestamp), spanning thousands of machines to petabyte scale; ancestor of wide-column NoSQL and the LSM engine
Chang et al. · 2006
Paper 4
The Chubby Lock Service — wraps error-prone Paxos into a "files + coarse-grained locks" service where whoever grabs the lock is master; the shared ancestor of GFS/Bigtable leader election and of ZooKeeper/etcd
Burrows · 2006
Paper 5
Percolator (Caffeine) — cross-row transactions with primary-lock two-phase commit plus observer notifications on Bigtable, turning Google's search index from full rebuilds into incremental updates and doubling freshness
Peng & Dabek · 2010
Paper 6
Pregel — a vertex-centric, BSP model for graph processing: write only "what one vertex does in one round," and the system handles partitioning, synchronization and fault tolerance at billion-vertex scale
Malewicz et al. · 2010
Paper 7
Dremel — columnar storage for nested data (repetition/definition levels shred and reassemble losslessly) plus a search-engine-style serving tree over thousands of machines, for second-scale interactive aggregation over trillions of rows; the basis of BigQuery
Melnik et al. · 2010
Paper 8
Spanner — each clock honestly reports "the time is somewhere in this small interval" (TrueTime), and a short commit-wait waits the error out, so a planet-spanning database gains a transaction total order everyone agrees on (external consistency)
Corbett et al. · 2012
Paper 9
F1 — a distributed SQL layer built on Spanner: it re-fits a "scalable + strongly consistent" storage foundation with full SQL, transactions, and consistent indexes, spreading cross-data-center latency thin via hierarchical clustering and optimistic transactions; it ran AdWords
Shute et al. · 2013
Paper 10
Dapper — give each request a trace id that travels with it, record each unit of work as a span, reassemble a call tree; transparent via shared-library probes, near-zero overhead via sampling and out-of-band collection — the blueprint for distributed tracing and observability
Sigelman et al. · 2010
Paper 11
The Tail at Scale — you wait on the slowest machine, so rare slowness gets amplified by scale; tolerate the latency tail the way we tolerate faults
Dean & Barroso · 2013
Consensus · Consistency · Storage
— clocks / consensus / NoSQL / P2P
Paper 1
Time, Clocks, and the Ordering of Events — with no clock everyone trusts, define order by "a message is sent before it's received," give each machine an ever-increasing ticket number, and machines a world apart produce one ordering everyone agrees on
Lamport · 1978
Paper 2
The Byzantine Generals Problem — reaching agreement when nodes may lie and send conflicting messages: solvable only if traitors are under one-third, via relaying plus majority vote; with unforgeable signatures any number of traitors is tolerable — the origin of every anti-cheating consensus
Lamport, Shostak & Pease · 1982
Paxos Made Simple — make crash-prone machines that only talk over an unreliable network agree on one value, uniquely and irreversibly: draw an ever-larger ticket, ask a majority first, honor the bigger ticket, and reuse any already-accepted value before pushing your own; since any two majorities overlap, a chosen value can never be overturned
Lamport · 2001
In Search of an Understandable Consensus Algorithm (Raft) — rewrote a consensus algorithm no one had grasped for thirty years into rules you can read and implement correctly: single leader + randomized elections + commit-on-majority, now the consistency core of etcd, CockroachDB, and countless systems
Ongaro & Ousterhout · 2014
Dynamo: Amazon's Highly Available Key-value Store — built so the cart never refuses a write: it trades immediate consistency for perpetual availability, using consistent hashing for placement, a tunable R+W quorum, and vector clocks that hand version conflicts to the app — igniting a whole generation of eventually consistent NoSQL like Cassandra
DeCandia et al. · 2007
Harvest, Yield, and Scalable Tolerant Systems — split availability into two dials, yield and harvest, for graceful degradation — the idea that became CAP
Fox & Brewer · 1999
Chord: A Scalable Peer-to-peer Lookup Service — turns "among millions of nodes coming and going, who owns this key" into a provable minimal primitive: nodes and keys sit on a consistent-hashing ring, a key goes to the nearest node clockwise, a "distance-doubling" finger table locates it in O(log N) hops, and a stabilization protocol self-heals — bedrock of a generation of NoSQL and P2P systems
Stoica et al. · 2001
Kafka — remade the message queue as an append-only partitioned log: the broker tracks no consumption and deletes nothing on read, so many parties share one stream at their own pace and can rewind
Kreps et al. · 2011
Databases · Data Models
Paper 1
A Relational Model of Data — lay data out as tidy tables linked by shared values rather than physical pointers, turning queries from "teach it how to walk" into "state what you want"; the theoretical root of nearly every modern database and of SQL
E. F. Codd · 1970
Operating Systems · Networking
Paper 1
The UNIX Time-Sharing System — a few minimal ideas (everything is a file, small tools joined by pipes, a tree of folders) build a small yet general OS; Linux, macOS, Android, and iOS are all its descendants
Ritchie & Thompson · 1974
Paper 2
End-to-End Arguments in System Design — some functions only the two ends can get right (like confirming a file arrived byte-for-byte); the network in the middle can't do them completely and drags everyone down trying, so keep the clever work at the ends and the network simple — the intellectual foundation of TCP/IP's "dumb pipes, smart endpoints"
Saltzer, Reed & Clark · 1984
Paper 3
Congestion Avoidance and Control — with no central dispatcher, let each sender pace its packets to the returning echoes (ACKs), nudge up while things go well, and halve on any loss; this shared "nudge up, halve down" self-discipline keeps the internet from collapsing together under load
Van Jacobson · 1988
Cryptography · Security · Decentralization
New Directions in Cryptography — split the key into a public and private half so two strangers can agree on a secret in the open; the bedrock of modern encryption
Diffie & Hellman · 1976
RSA — the first practical public-key scheme, betting encryption and digital signatures on the hardness of factoring large numbers
Rivest, Shamir & Adleman · 1978
Bitcoin (A Peer-to-Peer Electronic Cash System) — "solve a hard puzzle to record + longest chain wins + reward the recorder in coin" lets anonymous strangers stop double-spending with no bank, the origin of the blockchain
Satoshi Nakamoto · 2008
Reflections on Trusting Trust — hide a backdoor inside the "compiler compiles itself" step so it self-reproduces and leaves not one word in any source, proving "you can't trust code you did not totally create yourself," the intellectual origin of the supply-chain attack
Ken Thompson · 1984
Web · Search · Information
As We May Think — in 1945 imagined a desk called the Memex that ties knowledge into named, shareable "trails" by association, the intellectual source of the hyperlink and the Web
Vannevar Bush · 1945
The Anatomy of a Search Engine — treat every link as a weighted recommendation (PageRank) and describe a page by "what others call it" (anchor text) to surface the page most worth seeing in an exploding web, where Google began
Brin & Page · 1998
Foundations of Computing & Information
A Mathematical Theory of Communication — weighing information in "bits" and proving two iron laws: every source has a compression floor (entropy), every channel a speed limit (capacity) under which noise can be corrected to near-zero error
Claude Shannon · 1948
On Computable Numbers (The Turing Machine) — defining "computing" as a minimal tape machine, building the universal machine that "reads an instruction sheet to become any machine," then proving some problems no machine can ever decide
Alan Turing · 1936
Computing Machinery and Intelligence (The Turing Test) — replacing the unanswerable "can machines think?" with an imitation game: if a judge chatting by teleprinter can't tell machine from human, call it thinking — the founding text of AI philosophy and its first touchstone
Alan Turing · 1950
The Complexity of Theorem-Proving Procedures (NP-completeness) — making "hard to solve" precise: proving SAT the "hardest representative" so cracking it cracks a whole class of "easy to check, brutal to solve" problems, turning P vs NP into an exact million-dollar open question
Stephen Cook · 1971
Software Engineering · Programming
Go To Statement Considered Harmful — a two-page letter arguing goto unmoors "where you read" from "where the run is"; keep only sequence/branch/loop, and structured programming is born
Edsger Dijkstra · 1968
On the Criteria To Be Used in Decomposing Systems into Modules — split modules by the design decisions most likely to change, not by processing step; the source of encapsulation and OOP
David Parnas · 1972
No Silver Bullet: Essence and Accident in Software Engineering — software's difficulty splits into essence and accident; past advances only shaved accident, now nearly spent, so no bullet gives a 10× gain
Fred Brooks · 1986