Methods › General › Attention Mechanisms › FAVOR+

Fast Attention Via Positive Orthogonal Random Features

FAVOR+

102 papers tagged archive 2025-07-28

archive 2025-07-28 Description, source and code snippet are the archive's method entry.

FAVOR+, or Fast Attention Via Positive Orthogonal Random Features, is an efficient attention mechanism used in the Performer architecture which leverages approaches such as kernel methods and random features approximation for approximating softmax and Gaussian kernels.

FAVOR+ works for attention blocks using matrices A ∈ℝ^(L×L) of the form A(i, j) = K(qᵢᵀ, kⱼᵀ), with qᵢ/kⱼ standing for the iᵗʰ/jᵗʰ query/key row-vector in Q/K and kernel K : ℝᵈ × ℝᵈ →ℝ₊ defined for the (usually randomized) mapping: ϕ: ℝᵈ → ℝʳ₊ (for some r > 0) as:

K(x, y) = E[ϕ(x)ᵀϕ(y)]

We call ϕ(u) a random feature map for u ∈ℝᵈ . For Q^’, K^’ ∈ℝ^(L ×r) with rows given as ϕ(qᵢᵀ)ᵀ and ϕ(kᵢᵀ)ᵀ respectively, this leads directly to the efficient attention mechanism of the form:

Ât̂t̂_̂↔̂(Q, K, V) = D̂⁻¹(Q^’((K^’)ᵀV))

where

D̂ = diag(Q^’((K^’)1_L))

The above scheme constitutes the FA-part of the FAVOR+ mechanism. The other parts are achieved by:

The details are quite technical, so it is recommended you read the paper for further information on these steps.

Source: Rethinking Attention with Performers

In Syntology repository known, no verified implementation (declared_code; repositories: 5 listed of 7 in the graph, strongest link: official). The archive credits no paper with an arXiv id for this method, so whether this answer (about arXiv:2009.14794) is about this method is not checked.

Code Syntology holds under other names, as the lookup lists them (code_nearby); not a complete list, and none of them is this method.

Origin paper of this answer, arXiv:2009.14794 (its page on this site · its page on Syntology): Syntology's run record for its harvested code, 11 ran (of which 3 constructed an object rather than computing a result; 7 with no instrument failure: 1 honoured, 0 violated, 6 with no contract checked; 4 where Syntology's instrument failed) · 5 unverified (of 16 harvested samples) · 6 pointer-only (licence). The instrument figure counts failures of Syntology's instrument, not of the code. Ran means executed on a synthesized input, not that the code is correct.

Syntology's have check for this name (free, it answers coverage only), read 2026-10-05; ask it again live. A level is the served verification level of that one answer; nothing here is summed across methods.

What each level means, quoted from Syntology's server card: V0 “imports”; V1 “executes against a synthesized fixture and returns a value of the recorded type”; V2 “two implementations GENERATED from the paper's signature and spec agree numerically (1e-6) on fixture inputs” “neither of the two ever read the authors' own code, so V2 measures how unambiguously the paper specifies the method, not fidelity to the authors' implementation (measured 2026-09-11, commit 75eaa77: cross_checked_by is a second LLM on 2,628 of 2,831 served samples)”; V3 “paper-derived property tests pass on both of those same two generations.” The card adds: “No rung on this ladder compares the served code against the authors' implementation, so treat every level as evidence about runnability and internal agreement, not about matching the paper's results.” A level quoted in two parts is one definition the card writes as two clauses joined by a double hyphen, each clause quoted on its own (how these levels are quoted).

Papers archive 2025-07-28

archive papers tagged: 102 · with a code link: 43 · where Syntology ran a sample: 9 (8 with a run with no instrument failure, 1 where every run was a failure of Syntology's instrument) Syntology

Show: all tagged papersonly where code ran (9 of 102 tagged: 8 with a run with no instrument failure, 1 where every run was a failure of Syntology's instrument)

Page 1 of 2: papers 1 to 100 of 102, newest first by the archive's date (ties by slug), in archive order.

Repository counts are the archive's code-links table. A Syntology line states what Syntology ran from that paper's harvested code, as “N ran (of which C constructed an object rather than computing a result; K with no instrument failure: H honoured, V violated, P with no contract checked; I where Syntology's instrument failed) · U unverified”; the instrument figure counts failures of Syntology's instrument, not of the code. It is per sample and not a correctness claim. When the archive marks a repository official for the paper, the line starts with that repository's state (the archive's flag, not a verdict on who wrote the code; “community repositories only” when every sample that ran came from a community repository, “official: no sample here; runs from other or unrecorded repositories” when some came from a repository the paper names or has in its text, or from none recorded); hover it for the repositories the samples that ran came from.

Tasks archive 2025-07-28

20 shown of 143 tasks the archive attaches to papers tagged with this method, by distinct papers. A task without a page in the catalog is plain text.

TaskPapers
Language Modelling6
Decoder5
Time Series5
Classification4
Language Modeling4
Time Series Analysis4
Anomaly Detection3
Computational Efficiency3
Decision Making3
GPU3
Image Classification3
NeRF3
Novel View Synthesis3
Segmentation3
Sentiment Analysis3
image-classification3
Autonomous Driving2
Benchmarking2
Clustering2
Deep Reinforcement Learning2

Usage over time archive 2025-07-28

Papers per year tagged with FAVOR+: 2020 to 2025, peak 25 25 0 2020: 3 papers 2020 2021: 17 papers 2021 2022: 23 papers 2022 2023: 23 papers 2023 2024: 25 papers 2024 2025: 11 papers 2025
Papers per year the archive tags with this method, by the paper's archive date (102 dated). Bars are counts, not a trend claim.

Components: the archive holds no method-to-method composition, so PwC's Components table cannot be rebuilt; the Papers list carries no Results column for the same reason (the archive does not join its leaderboard rows to method tags).

Categories archive 2025-07-28

Attention Mechanisms

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections