arXiv subjects

Carl Kingsford

Publications and source records attributed to Carl Kingsford.

17 recordsLinked to original sources

arXiv2026

The Memory Curse: How Expanded Recall Erodes Cooperative Intent in LLM Agents

Context window expansion is often treated as a straightforward capability upgrade for LLMs, but we find it systematically fails in multi-agent social dilemmas. Across 7 LLMs and 4 games over 500 rounds, expanding accessible history degrades cooperation in 18 of 28 model--game settings, a pattern we term the memory curse. We isolate the underlying mechanism through three analyses. First, lexical analysis of 378,000 reasoning traces associates this breakdown with eroding forward-looking intent rather than rising paranoia. We validate this using targeted fine-tuning as a cognitive probe: a LoRA adapter trained exclusively on forward-looking traces mitigates the decay and transfers zero-shot to distinct games. Second, memory sanitization holds prompt length fixed while replacing visible history with synthetic cooperative records, which restores cooperation substantially, proving the trigger is memory content, not length alone. Finally, ablating explicit Chain-of-Thought reasoning often reduces the collapse, showing that deliberation paradoxically amplifies the memory curse. Together, these results recast memory as an active determinant of multi-agent behavior: longer recall can either destabilize or support cooperation depending on the reasoning patterns it elicits.

cs.CL↗

arXiv2026

Why Search When You Can Transfer? Amortized Agentic Workflow Design from Structural Priors

Automated agentic workflow design currently relies on per-task iterative search, which is computationally prohibitive and fails to reuse structural knowledge across tasks. We observe that optimized workflows converge to a small family of domain-specific topologies, suggesting that this combinatorial search is largely redundant. Building on this insight, we propose SWIFT (Synthesizing Workflows via Few-shot Transfer), a framework that amortizes workflow design into reusable structural priors. SWIFT first distills compositional heuristics and output-interface contracts from contrastive analysis of prior search trajectories across source tasks. At inference time, it conditions a single LLM generation pass on these priors together with cross-task workflow demonstrations to synthesize a complete, executable workflow for an unseen target task, bypassing iterative search entirely. On five benchmarks, SWIFT outperforms the state-of-the-art search-based method while reducing marginal per-task optimization cost by three orders of magnitude. It further generalizes to four additional unseen benchmarks and transfers successfully from GPT-4o-mini to three additional foundation models (Grok, Qwen, Gemma). Controlled ablations reveal that workflow demonstrations primarily transfer topological structure rather than surface semantics: replacing all operator names with random strings still retains over 93% of the full system's average performance.

cs.LG↗

arXiv2026

Models Know Their Shortcuts: Deployment-Time Shortcut Mitigation

Pretrained text encoders are prone to shortcut learning, relying on token-label correlations that fail once the distribution shifts in deployment. Existing shortcut mitigation methods mainly operate at training time and assume access to training data, training dynamics, or shortcut annotations, which are hardly available during deployment, where only the converged model remains. We show that this model alone suffices to mitigate shortcuts during deployment: a biased model internalizes a signal of its learned shortcuts that can be captured via unsupervised gradient-based attribution. We further prove that deployment-time mitigation is information-theoretically upper-bounded by training-time mitigation. Nevertheless, exploiting this gradient signal, our proposed unsupervised deployment-time shortcut mitigation framework for pretrained text encoders, Shortcut Guardrail, recovers substantial performance under shortcut distribution shift, matching or outperforming training-time baselines across sentiment classification, toxicity detection, and natural language inference.

cs.LG↗

arXiv2026

Turnpike with Uncertain Measurements: Triangle-Equality ILP with a Deterministic Recovery Guarantee

We study Turnpike with uncertain measurements: reconstructing a one-dimensional point set from an unlabeled multiset of pairwise distances under bounded noise and rounding. We give a combinatorial characterization of realizability via a multi-matching that labels interval indices by distinct distance values while satisfying all triangle equalities. This yields an ILP based on the triangle equality whose constraint structure depends only on the two-partition set $\mathcal{P}_y=\{(r,s,t): y_r+y_s=y_t\}$ and a natural LP relaxation with $\{0,1\}$-coefficient constraints. Integral solutions certify realizability and output an explicit assignment matrix, enabling an assignment-first, regression-second pipeline for downstream coordinate estimation. Under bounded noise followed by rounding, we prove a deterministic separation condition under which $\mathcal{P}_y$ is recovered exactly, so the ILP/LP receives the same combinatorial input as in the noiseless case. Experiments illustrate integrality behavior and degradation outside the provable regime.

cs.CG↗

arXiv2025

CodonMoE: DNA Language Models for mRNA Analyses

Genomic language models (gLMs) face a fundamental efficiency challenge: either maintain separate specialized models for each biological modality (DNA and RNA) or develop large multi-modal architectures. Both approaches impose significant computational burdens - modality-specific models require redundant infrastructure despite inherent biological connections, while multi-modal architectures demand massive parameter counts and extensive cross-modality pretraining. To address this limitation, we introduce CodonMoE (Adaptive Mixture of Codon Reformative Experts), a lightweight adapter that transforms DNA language models into effective RNA analyzers without RNA-specific pretraining. Our theoretical analysis establishes CodonMoE as a universal approximator at the codon level, capable of mapping arbitrary functions from codon sequences to RNA properties given sufficient expert capacity. Across four RNA prediction tasks spanning stability, expression, and regulation, DNA models augmented with CodonMoE significantly outperform their unmodified counterparts, with HyenaDNA+CodonMoE series achieving state-of-the-art results using 80% fewer parameters than specialized RNA models. By maintaining sub-quadratic complexity while achieving superior performance, our approach provides a principled path toward unifying genomic language modeling, leveraging more abundant DNA data and reducing computational overhead while preserving modality-specific performance advantages.

q-bio.GN↗

arXiv2024

Computationally Efficient High-Dimensional Bayesian Optimization via Variable Selection

Bayesian Optimization (BO) is a method for globally optimizing black-box functions. While BO has been successfully applied to many scenarios, developing effective BO algorithms that scale to functions with high-dimensional domains is still a challenge. Optimizing such functions by vanilla BO is extremely time-consuming. Alternative strategies for high-dimensional BO that are based on the idea of embedding the high-dimensional space to the one with low dimension are sensitive to the choice of the embedding dimension, which needs to be pre-specified. We develop a new computationally efficient high-dimensional BO method that exploits variable selection. Our method is able to automatically learn axis-aligned sub-spaces, i.e. spaces containing selected variables, without the demand of any pre-specified hyperparameters. We theoretically analyze the computational complexity of our algorithm and derive the regret bound. We empirically show the efficacy of our method on several synthetic and real problems.

cs.LG↗

arXiv2023

Revisiting the Complexity of and Algorithms for the Graph Traversal Edit Distance and Its Variants

The graph traversal edit distance (GTED), introduced by Ebrahimpour Boroojeny et al.~(2018), is an elegant distance measure defined as the minimum edit distance between strings reconstructed from Eulerian trails in two edge-labeled graphs. GTED can be used to infer evolutionary relationships between species by comparing de Bruijn graphs directly without the computationally costly and error-prone process of genome assembly. Ebrahimpour Boroojeny et al.~(2018) propose two ILP formulations for GTED and claim that GTED is polynomially solvable because the linear programming relaxation of one of the ILPs always yields optimal integer solutions. The claim that GTED is polynomially solvable is contradictory to the complexity results of existing string-to-graph matching problems. We resolve this conflict in complexity results by proving that GTED is NP-complete and showing that the ILPs proposed by Ebrahimpour Boroojeny et al. do not solve GTED but instead solve for a lower bound of GTED and are not solvable in polynomial time. In addition, we provide the first two, correct ILP formulations of GTED and evaluate their empirical efficiency. These results provide solid algorithmic foundations for comparing genome graphs and point to the direction of heuristics. The source code to reproduce experimental results is available at https://github.com/Kingsford-Group/gtednewilp/.

cs.DM↗

arXiv2023

Sketching methods with small window guarantee using minimum decycling sets

Most sequence sketching methods work by selecting specific $k$-mers from sequences so that the similarity between two sequences can be estimated using only the sketches. Estimating sequence similarity is much faster using sketches than using sequence alignment, hence sketching methods are used to reduce the computational requirements of computational biology software packages. Applications using sketches often rely on properties of the $k$-mer selection procedure to ensure that using a sketch does not degrade the quality of the results compared with using sequence alignment. In particular the window guarantee ensures that no long region of the sequence goes unrepresented in the sketch. A sketching method with a window guarantee corresponds to a Decycling Set, aka an unavoidable sets of $k$-mers. Any long enough sequence must contain a $k$-mer from any decycling set (hence, it is unavoidable). Conversely, a decycling set defines a sketching method by selecting the $k$-mers from the set. Although current methods use one of a small number of sketching method families, the space of decycling sets is much larger, and largely unexplored. Finding decycling sets with desirable characteristics is a promising approach to discovering new sketching methods with improved performance (e.g., with small window guarantee). The Minimum Decycling Sets (MDSs) are of particular interest because of their small size. Only two algorithms, by Mykkeltveit and Champarnaud, are known to generate two particular MDSs, although there is a vast number of alternative MDSs. We provide a simple method that allows one to explore the space of MDSs and to find sets optimized for desirable properties. We give evidence that the Mykkeltveit sets are close to optimal regarding one particular property, the remaining path length.

cs.DS↗

arXiv2021

How much data is sufficient to learn high-performing algorithms? Generalization guarantees for data-driven algorithm design

Algorithms often have tunable parameters that impact performance metrics such as runtime and solution quality. For many algorithms used in practice, no parameter settings admit meaningful worst-case bounds, so the parameters are made available for the user to tune. Alternatively, parameters may be tuned implicitly within the proof of a worst-case approximation ratio or runtime bound. Worst-case instances, however, may be rare or nonexistent in practice. A growing body of research has demonstrated that data-driven algorithm design can lead to significant improvements in performance. This approach uses a training set of problem instances sampled from an unknown, application-specific distribution and returns a parameter setting with strong average performance on the training set. We provide a broadly applicable theory for deriving generalization guarantees that bound the difference between the algorithm's average performance over the training set and its expected performance. Our results apply no matter how the parameters are tuned, be it via an automated or manual approach. The challenge is that for many types of algorithms, performance is a volatile function of the parameters: slightly perturbing the parameters can cause large changes in behavior. Prior research has proved generalization bounds by employing case-by-case analyses of greedy algorithms, clustering algorithms, integer programming algorithms, and selling mechanisms. We uncover a unifying structure which we use to prove extremely general guarantees, yet we recover the bounds from prior research. Our guarantees apply whenever an algorithm's performance is a piecewise-constant, -linear, or -- more generally -- piecewise-structured function of its parameters. Our theory also implies novel bounds for voting mechanisms and dynamic programming algorithms from computational biology.

cs.LG↗

arXiv2020

Lower density selection schemes via small universal hitting sets with short remaining path length

Universal hitting sets are sets of words that are unavoidable: every long enough sequence is hit by the set (i.e., it contains a word from the set). There is a tight relationship between universal hitting sets and minimizers schemes, where minimizers schemes with low density (i.e., efficient schemes) correspond to universal hitting sets of small size. Local schemes are a generalization of minimizers schemes which can be used as replacement for minimizers scheme with the possibility of being much more efficient. We establish the link between efficient local schemes and the minimum length of a string that must be hit by a universal hitting set. We give bounds for the remaining path length of the Mykkeltveit universal hitting set. Additionally, we create a local scheme with the lowest known density that is only a log factor away from the theoretical lower bound.

cs.DS↗

arXiv2016

Efficient Index Maintenance Under Dynamic Genome Modification

Efficient text indexing data structures have enabled large-scale genomic sequence analysis and are used to help solve problems ranging from assembly to read mapping. However, these data structures typically assume that the underlying reference text is static and will not change over the course of the queries being made. Some progress has been made in exploring how certain text indices, like the suffix array, may be updated, rather than rebuilt from scratch, when the underlying reference changes. Yet, these update operations can be complex in practice, difficult to implement, and give fairly pessimistic worst-case bounds. We present a novel data structure, SkipPatch, for maintaining a k-mer-based index over a dynamically changing genome. SkipPatch pairs a hash-based k-mer index with an indexable skip list that is used to efficiently maintain the set of edits that have been applied to the original genome. SkipPatch is practically fast, significantly outperforming the dynamic extended suffix array in terms of update and query speed.

q-bio.GN↗

arXiv2015

Optimal Seed Solver: Optimizing Seed Selection in Read Mapping

Motivation: Optimizing seed selection is an important problem in read mapping. The number of non-overlapping seeds a mapper selects determines the sensitivity of the mapper while the total frequency of all selected seeds determines the speed of the mapper. Modern seed-and-extend mappers usually select seeds with either an equal and fixed-length scheme or with an inflexible placement scheme, both of which limit the potential of the mapper to select less frequent seeds to speed up the mapping process. Therefore, it is crucial to develop a new algorithm that can adjust both the individual seed length and the seed placement, as well as derive less frequent seeds. Results: We present the Optimal Seed Solver (OSS), a dynamic programming algorithm that discovers the least frequently-occurring set of x seeds in an L-bp read in $O(x \times L)$ operations on average and in $O(x \times L^{2})$ operations in the worst case. We compared OSS against four state-of-the-art seed selection schemes and observed that OSS provides a 3-fold reduction of average seed frequency over the best previous seed selection optimizations.

cs.CE↗

arXiv2013

Sailfish: Alignment-free Isoform Quantification from RNA-seq Reads using Lightweight Algorithms

RNA-seq has rapidly become the de facto technique to measure gene expression. However, the time required for analysis has not kept up with the pace of data generation. Here we introduce Sailfish, a novel computational method for quantifying the abundance of previously annotated RNA isoforms from RNA-seq data. Sailfish entirely avoids mapping reads, which is a time-consuming step in all current methods. Sailfish provides quantification estimates much faster than existing approaches (typically 20-times faster) without loss of accuracy.

q-bio.GN↗

arXiv2013

Multiscale Identification of Topological Domains in Chromatin

Recent chromosome conformation capture experiments have led to the discovery of dense, contiguous, megabase-sized topological domains that are similar across cell types and conserved across species. These domains are strongly correlated with a number of chromatin markers and have since been included in a number of analyses. However, functionally-relevant domains may exist at multiple length scales. We introduce a new and efficient algorithm that is able to capture persistent domains across various resolutions by adjusting a single scale parameter. The identified novel domains are substantially different from domains reported previously and are highly enriched for insulating factor CTCF binding and histone modfications at the boundaries.

q-bio.QM↗

arXiv2010

Network Archaeology: Uncovering Ancient Networks from Present-day Interactions

Often questions arise about old or extinct networks. What proteins interacted in a long-extinct ancestor species of yeast? Who were the central players in the Last.fm social network 3 years ago? Our ability to answer such questions has been limited by the unavailability of past versions of networks. To overcome these limitations, we propose several algorithms for reconstructing a network's history of growth given only the network as it exists today and a generative model by which the network is believed to have evolved. Our likelihood-based method finds a probable previous state of the network by reversing the forward growth model. This approach retains node identities so that the history of individual nodes can be tracked. We apply these algorithms to uncover older, non-extant biological and social networks believed to have grown via several models, including duplication-mutation with complementarity, forest fire, and preferential attachment. Through experiments on both synthetic and real-world data, we find that our algorithms can estimate node arrival times, identify anchor nodes from which new nodes copy links, and can reveal significant features of networks that have long since disappeared.

q-bio.MN↗

arXiv2009

A synthesis for exactly 3-edge-connected graphs

A multigraph is exactly k-edge-connected if there are exactly k edge-disjoint paths between any pair of vertices. We characterize the class of exactly 3-edge-connected graphs, giving a synthesis involving two operations by which every exactly 3-edge-connected multigraph can be generated. Slightly modified syntheses give the planar exactly 3-edge-connected graphs and the exactly 3-edge-connected graphs with the fewest possible edges.

math.CO↗

arXiv2009

Vertices of degree k in edge-minimal, k-edge-connected graphs

Halin showed that every edge minimal, k-vertex connected graph has a vertex of degree k. In this note, we prove the analogue to Halin's theorem for edge-minimal, k-edge-connected graphs. We show there are two vertices of degree k in every edge-minimal, k-edge-connected graph.

math.CO↗