CS Paper Radar

3SUM Is Really Hard: A Real-to-Integer Reduction

Nick Fischer, Adam Polak, Jonas Schmidt

cs.DS ·submitted 2026-10-07 14:29 UTC · v1
top venuetheory

Abstract

We show that if the 3SUM problem on integer-valued inputs can be solved in truly subquadratic time, then it can also be solved in truly subquadratic time on real-valued inputs. This answers an open problem posed by Chan, Vassilevska Williams, and Xu [STOC 2022], and constitutes the first such tight real-to-integer self-reduction in fine-grained complexity. Our proof relies on a surprising combination of the Frank--Tardos theorem from optimization with Freiman-type theorems from additive combinatorics.

Comment
To appear in FOCS 2026

Closest papers

Nearest neighbours by meaning across the whole index.

Real-to-Integer Reduction for All-Pairs Shortest Paths

cs.DS ·2026-11-03 ·Nick Fischer, Adam Polak, Jonas Schmidt
theory similarity 0.79
Abstract

Efficient graph algorithms often use techniques that are specific to integer-valued inputs and do not work for real-valued versions of the problems. Is it a limitation of current techniques, or are the real-valued variants inherently harder? One way to study this question is through real-to-integer reductions. We give (randomized, combinatorial) subcubic real-to-integer reductions for two central graph problems in fine-grained complexity: All-Pairs Shortest Paths and Exact Triangle. That is, we show that if the integer variants of these problems can be solved in truly subcubic time, then the real variants can also be solved in truly subcubic time. Our work extends and combines techniques from three previous works: (1) our real-to-integer reduction for the 3SUM problem (Fischer, Polak, Schmidt, FOCS 2026), which itself combines the Frank--Tardos theorem from optimization with Freiman's theorem from additive combinatorics; (2) universe reduction for APSP (Fischer, STOC 2026), which we generalize to reals and use to reduce the doubling of the set of edge weights; and (3) the ``Fredman's trick meets dominance product'' technique (Chan, Vassilevska Williams, Xu, STOC 2023), which is needed to adapt the prior real-to-integer framework from equality constraints to inequalities. As a corollary, our reduction combined with Yuval's fast Integer APSP algorithm in the real RAM model with multiplication (IPL 1976) shows that the Real APSP Hypothesis is false in this model, which also follows from the very recent breakthrough by Alman and Vassilevska Williams using a very different approach.

0.05

Preprocessed 3SUM: From Reals to Integers

cs.DS ·2026-10-15 ·Nick Fischer, Jonas Schmidt
similarity 0.67
Abstract

The Preprocessed 3SUM problem asks to preprocess a set $A$ of $n$ numbers so that we can quickly decide whether a query subset $A' \subseteq A$ contains a solution $a+b=c$. The best known query bounds differ substantially between integer and real inputs. For integers, a series of advances with near-quadratic preprocessing reduced the query time from $\tilde O(n^{13/7})$ [Chan, Lewenstein; STOC '15] to $\tilde O(n^{11/6})$ [Chan, Vassilevska W., Xu; STOC '23], and finally to $\tilde O(n^{3/2})$ [Kasliwal, Polak, Sharma; SOSA '25]. For reals, only the first of these bounds has been achieved, by adapting the corresponding integer algorithm [Fischer; SODA '25]. Is this gap inherent to the more limited tools available for real inputs? We give a randomized real-to-integer reduction with $\tilde O(n^2)$ preprocessing overhead and no asymptotic increase in query time. This yields $\tilde O(n^2)$ preprocessing and $\tilde O(n^{3/2})$ query time for real inputs, closing the gap. We show how to extend our result to the setting where one set is unknown at preprocessing time. Our reduction is conceptually simple, and relies on an interesting technical toolkit around old and new tools from sparse linear algebra. More broadly, it suggests a route to real-to-integer reductions for other additive problems.

0.34

Pigeonhole Equal Subset Sum: Subexponential Algorithm, Tight Lower Bounds, and Average-Case Analysis

cs.DS ·2026-10-22 ·Deepak Bhati, Pranjal Dutta, Antoine Joux, Mahesh Sreekumar Rajasree, Karol Wegrzycki
top venuetheory similarity 0.62
Abstract · 66 pages, 2 figures, published in FOCS 2026

The Equal Subset Sum problem asks, given n positive integers bounded by M, to find two distinct non-empty subsets with the same sum. We study this problem in the regime $M = 2^{o(n)}$, where the pigeonhole principle guarantees that a solution exists, and the main challenge is to find one efficiently. A straightforward dynamic-programming algorithm runs in time poly(n).$M$. After a sequence of improvements, the current best algorithm, due to Jin, Williams, and Zhang (ESA 2025), achieves running time poly(n).$M^{1/3}$. In this work, we give the first poly(n).$M^{o(1)}$-time algorithm for Equal Subset Sum in the regime $M = 2^{o(n)}$. We complement this algorithmic result with a tight conditional lower bound showing that any substantial improvement would require a major breakthrough in approximation algorithms for lattice problems, building on recent techniques of Vafa and Vaikuntanathan (STOC 2025). Finally, we show that our techniques yield an asymptotically faster algorithm for the regime when $M = Θ(2^{n})$: a poly(n)$\cdot 2^{n/4}$-time algorithm in the average case.

0.32

A simplified proof of polynomial improvements for 3SUM and APSP

cs.CC ·2026-10-11 ·Sankeerth Rao Karingula, Shachar Lovett
publishedhot directionhighly cited similarity 0.61
Abstract

We give a simplified proof of the polynomial improvements for integer 3SUM and all-pairs shortest paths (APSP) established by Alman and Vassilevska Williams, with weaker exponents. The central ingredient is selected-entry matrix multiplication: computing a prescribed set of entries of a product of dense rectangular matrices. Starting from Schönhage's inner/outer-product identity, we construct a multiplication rule that shares intermediate products among many outputs. We apply this rule recursively and bound the work by counting the requested entries at each level. Applied to incidence matrices, the algorithm solves Offline Set Disjointness over a small universe. Known deterministic reductions then yield truly subquadratic integer 3SUM and truly subcubic APSP for polynomially bounded integer inputs.

0.58

Beyond 3SUM-Hardness for Geometric Problems

cs.DS ·2026-10-23 ·Nick Fischer
similarity 0.60
Abstract

Since the early days of fine-grained complexity, many geometric problems have been shown to require quadratic time under the 3SUM Hypothesis. But as the 3SUM Hypothesis was recently refuted, these lower bounds have lost their foundation. In this note we point out that many of them can be recovered under an alternative hardness assumption, the $k$-AP Hypothesis. The required changes to the proofs are trivial, so the lower bounds remain just as easy to teach.

0.39

Additive $Ω(\log^* OPT)$ Gap for Bin Packing with Three Items per Bin

cs.DS ·2026-11-06 ·Ilan Doron-Arad
theory similarity 0.57
Abstract

We prove that bin packing is NP-hard to approximate within an additive factor of $o(\ls\OPT)$, even when all item sizes are arbitrarily close to $1/3$. This gives the first lower bound for bin packing with an additive factor that depends explicitly on the optimum value. In addition, we give a polynomial-time reduction from any bin packing instance into one with item sizes arbitrarily close to $1/3$, preserving its additive configuration-LP gap up to constant factors. Thus, in the sense of additive configuration LP gap and up to a constant, we confirm the conjecture that 3-partition instances are the hardest bin packing instances. In particular, we obtain an unconditional $Ω(\ls\OPT)$ gap for 3-partition instances. We build on the recent OpenAI construction that uses five items per bin and rules out any constant additive gap.

0.25

On the Largest Cardinality Subset Sum

cs.DS ·2026-10-21 ·Deepak Bhati, Anubhav Dhar, Pranjal Dutta, Mahesh Sreekumar Rajasree, Karol Węgrzycki
similarity 0.57
Abstract · Submitted to STACS 2027

We study Largest Cardinality Subset Sum (LCSS): given $n$ nonnegative integers and a target $t$, compute a subset summing to $t$ with maximum cardinality. We also consider its all-targets version (ATLCSS), which asks for this maximum for every sum $s\in\{1,\ldots,t\}$. The LCSS problem sits squarely between the subset sum and the knapsack problem. A straightforward dynamic-programming algorithm runs in time $O(nt)$ and solves ATLCSS, and using Jin and Bringmann (both in STOC 2024, independent work) $0$-$1$ knapsack results we achieve a $\widetilde{O}(n + a_{\max}^2)$-time deterministic algorithm for LCSS. Our main result is a randomized $\widetilde{O}(n+\sqrt{n}\,t)$-time algorithm for LCSS. Then, we complement this result with a randomized $\widetilde{O}(n+t^{3/2})$-time algorithm for ATLCSS. Together, our two algorithms give a $\widetilde{O}(n+t\min\{\sqrt{n},\sqrt{t}\})$-time algorithm for LCSS. As an application of ATLCSS, we solve the search version of Projection Subset Sum, introduced by Dutta and Rajasree (SYNSAC 2023), in $\widetilde{O}(n+t^{3/2})$ time for every fixed $\ell_p$ norm with $p\geq 1$. For $\ell_1$, we obtain $\widetilde{O}(n+t)$ time, removing the unique-representation assumption required by their algorithm. Both algorithms output an optimal modified vector and a witnessing subset.

0.19

Dense Pinwheel Packing Is Strongly NP-Complete

cs.CC ·2026-09-17 ·Yusuke Kobayashi, Bingkai Lin, Joseph Swernofsky
theory similarity 0.57
Abstract

An instance of {\sc Pinwheel Packing} is a list of positive integers $a_1,\ldots,a_k$. A feasible schedule assigns one task to every integer time so that every interval of $a_i$ consecutive times contains task $i$. The instance is \emph{dense} when $\sum_i1/a_i=1$. We prove that {\sc Dense Pinwheel Packing} is NP-complete even when every period is encoded in unary and equal periods are listed as distinct tasks. Consequently, the usual binary-encoded problem is strongly NP-complete. Kleinberg and Mishra also prove NP-completeness \cite[Corollary~5.1]{KleinbergMishra2026}, but their reduction uses periods of exponential numerical size and therefore yields only weak NP-hardness. Our proof uses a direct reduction from triangle partition in a sparse tripartite graph. If each of the three parts of the source graph has $n$ vertices, the reduction produces $O(n^4\log^3 n)$ explicitly listed tasks, each with period $O(n^4\log^3 n)$; consequently, its full unary encoding has length $O(n^8\log^6 n)$.

0.33