Search

Citation intention Classification (CIC) tools classify citations by their intention (e.g., background, motivation) and assist readers in evaluating the contribution of scientific literature. Prior research has shown that pretrained language models (PLMs) such as SciBERT can achieve state-of-the-art performance on CIC benchmarks. PLMs are trained via self-supervision tasks on a large corpus of general text and can quickly adapt to CIC tasks via moderate fine-tuning on the corresponding dataset. Despite their advantages, PLMs can easily overfit small datasets during fine-tuning. In this paper, we propose a multi-task learning (MTL) framework that jointly fine-tunes PLMs on a dataset of primary interest together with multiple auxiliary CIC datasets to take advantage of additional supervision signals. We develop a data-driven task relation learning (TRL) method that controls the contribution of auxiliary datasets to avoid negative transfer and expensive hyper-parameter tuning. We conduct experiments on three CIC datasets and show that fine-tuning with additional datasets can improve the PLMs' generalization performance on the primary dataset. PLMs fine-tuned with our proposed framework outperform the current state-of-the-art models by 7% to 11% on small datasets while aligning with the best-performing model on a large dataset., Comment: To be appear as a Findings paper at EMNLP 2024

The ICPR 2024 Competition on Safe Segmentation of Drive Scenes in Unstructured Traffic and Adverse Weather Conditions served as a rigorous platform to evaluate and benchmark state-of-the-art semantic segmentation models under challenging conditions for autonomous driving. Over several months, participants were provided with the IDD-AW dataset, consisting of 5000 high-quality RGB-NIR image pairs, each annotated at the pixel level and captured under adverse weather conditions such as rain, fog, low light, and snow. A key aspect of the competition was the use and improvement of the Safe mean Intersection over Union (Safe mIoU) metric, designed to penalize unsafe incorrect predictions that could be overlooked by traditional mIoU. This innovative metric emphasized the importance of safety in developing autonomous driving systems. The competition showed significant advancements in the field, with participants demonstrating models that excelled in semantic segmentation and prioritized safety and robustness in unstructured and adverse conditions. The results of the competition set new benchmarks in the domain, highlighting the critical role of safety in deploying autonomous vehicles in real-world scenarios. The contributions from this competition are expected to drive further innovation in autonomous driving technology, addressing the critical challenges of operating in diverse and unpredictable environments., Comment: 15 pages, 7 figures, ICPR Competition Paper

The rapid progress in machine learning (ML) has brought forth many large language models (LLMs) that excel in various tasks and areas. These LLMs come with different abilities and costs in terms of computation or pricing. Since the demand for each query can vary, e.g., because of the queried domain or its complexity, defaulting to one LLM in an application is not usually the best choice, whether it is the biggest, priciest, or even the one with the best average test performance. Consequently, picking the right LLM that is both accurate and cost-effective for an application remains a challenge. In this paper, we introduce MetaLLM, a framework that dynamically and intelligently routes each query to the optimal LLM (among several available LLMs) for classification tasks, achieving significantly improved accuracy and cost-effectiveness. By framing the selection problem as a multi-armed bandit, MetaLLM balances prediction accuracy and cost efficiency under uncertainty. Our experiments, conducted on popular LLM platforms such as OpenAI's GPT models, Amazon's Titan, Anthropic's Claude, and Meta's LLaMa, showcase MetaLLM's efficacy in real-world scenarios, laying the groundwork for future extensions beyond classification tasks.

A core objective of physical design is to minimize wirelength (WL) when placing chip components on a canvas. Computing the minimal WL of a placement requires finding rectilinear Steiner minimum trees (RSMTs), an NP-hard problem. We propose NeuroSteiner, a neural model that distills GeoSteiner, an optimal RSMT solver, to navigate the cost--accuracy frontier of WL estimation. NeuroSteiner is trained on synthesized nets labeled by GeoSteiner, alleviating the need to train on real chip designs. Moreover, NeuroSteiner's differentiability allows to place by minimizing WL through gradient descent. On ISPD 2005 and 2019, NeuroSteiner can obtain 0.3% WL error while being 60% faster than GeoSteiner, or 0.2% and 30%., Comment: Work-in-Progress poster at the 2024 Design and Automation Conference (DAC'24)

We study the effect of network topology on the collective dynamics of an oscillator ensemble. Specifically, we explore explosive synchronization in a system of interacting star networks. Explosive synchronization is characterized by an abrupt transition from an incoherent state to a coherent state. In this study, we couple multiple star networks through their hubs and study the emergent dynamics as a function of coupling strength. The dynamics of each node satisfies the equation of a Kuramoto oscillator. We observe that for a small inter-star coupling strength, the hysteresis width between the forward and backward transition point is minimal, which increases with an increase in the inter-star coupling strength. This observation is independent of the size of the network. Further, we find that the backward transition point is independent of the number of stars coupled together and the inter-star coupling strength, which is also verified using the Watanabe and Strogatz (WS) theory., Comment: 8 pages, 5 figures

Enterprises grapple with the significant challenge of managing proprietary unstructured data, hindering efficient information retrieval. This has led to the emergence of AI-driven information retrieval solutions, designed to adeptly extract relevant insights to address employee inquiries. These solutions often leverage pre-trained embedding models and generative models as foundational components. While pre-trained embeddings may exhibit proximity or disparity based on their original training objectives, they might not fully align with the unique characteristics of enterprise-specific data, leading to suboptimal alignment with the retrieval goals of enterprise environments. In this paper, we propose a comprehensive methodology for contextualizing pre-trained embedding models to enterprise environments, covering the entire process from data preparation to model fine-tuning and evaluation. By adapting the embeddings to better suit the retrieval tasks prevalent in enterprises, we aim to enhance the performance of information retrieval solutions. We discuss the process of fine-tuning, its effect on retrieval accuracy, and the potential benefits for enterprise information management. Our findings demonstrate the efficacy of fine-tuned embedding models in improving the precision and relevance of search results in enterprise settings.

Internet of Things (IoT) has seen a prolific rise in recent times and provides the ability to solve several key challenges faced by our societies and environment. Data produced by IoT provides a significant opportunity to infer context that is key for IoT applications to make decisions/actuations. Context Management Platform (CMP) is a middleware to facilitate the exchange and management of such context information among IoT applications. In this paper, we propose a novel approach to monitoring context freshness as a key metric, to improving the CMP's caching performance to support the real-time context needs of IoT applications. Our proposed hybrid algorithm uses Analytic Hierarchy Process (AHP) and Sliding Window technique to ensure the most relevant (as needed by the IoT applications) context information is cached. By continuously monitoring and prioritizing context attributes, the strategy adapts to IoT environment changes, keeping cached context fresh and reliable. Through experimental evaluation and using mock data obtained from a real-world mobile IoT scenario in section~\ref{use case}, we demonstrate that the proposed algorithm can substantially enhance context cache performance, by monitoring the context attributes in real time.

Trapped materials at the interfaces of two-dimensional heterostructures (HS) lead to reduced coupling between the layers, resulting in degraded optoelectronic performance and device variability. Further, nanobubbles can form at the interface during transfer or after annealing. The question of what is inside a nanobubble, i.e. the trapped material, remains unanswered, limiting the studies and applications of these nanobubble systems. In this work, we report two key advances. Firstly, we quantify the interface quality using RAW-format optical imaging, and distinguish between ideal and non-ideal interfaces. The HS-substrate ratio value is calculated using a transfer matrix model, and is able to detect the presence of trapped layers. The second key advance is identification of water as the trapped material inside a nanobubble. To the best of our knowledge, this is the first study to show that optical imaging alone can quantify interface quality, and find the type of trapped material inside spontaneously formed nanobubbles. We also define a quality index parameter to quantify the interface quality of HS. Quantitative measurement of the interface will help answer the question whether annealing is necessary during HS preparation, and will enable creation of complex HS with small twist angles. Identification of the trapped materials will pave the way towards using nanobubbles for novel optical and engineering applications.

Lipidomics generates large data that makes manual annotation and interpretation challenging. Lipid chemical and structural diversity with structural isomers further complicates annotation. Although, several commercial and open-source software for targeted lipid identification exists, it lacks automated method generation workflows and integration with statistical and bioinformatics tools. We have developed the Comprehensive Lipidomic Automated Workflow (CLAW) platform with integrated workflow for parsing, detailed statistical analysis and lipid annotations based on custom multiple reaction monitoring (MRM) precursor and product ion pair transitions. CLAW contains several modules including identification of carbon-carbon double bond position(s) in unsaturated lipids when combined with ozone electrospray ionization (OzESI)-MRM methodology. To demonstrate the utility of the automated workflow in CLAW, large-scale lipidomics data was collected with traditional and OzESI-MRM profiling on biological and non-biological samples. Specifically, a total of 1497 transitions organized into 10 MRM-based mass spectrometry methods were used to profile lipid droplets isolated from different brain regions of 18-24 month-old Alzheimer's disease mice and age-matched wild-type controls. Additionally, triacyclglycerols (TGs) profiles with carbon-carbon double bond specificity were generated from canola oil samples using OzESI-MRM profiling. We also developed an integrated language user interface with large language models using artificially intelligent (AI) agents that permits users to interact with the CLAW platform using a chatbot terminal to perform statistical and bioinformatic analyses. We envision CLAW pipeline to be used in high-throughput lipid structural identification tasks aiding users to generate automated lipidomics workflows ranging from data acquisition to AI agent-based bioinformatic analysis., Comment: 53 pages, 4 main figures, 23 Supporting figures, 10 Supporting Tables

Constraint-based optimization of microbial strains and model-based bioprocess design have been used extensively to enhance yields in biotechnological processes. However, strain and process optimization are usually carried out in sequential steps, causing underperformance of the biotechnological process when scaling up to industrial fermentation conditions. Herein, we propose the optimization formulation SimulKnock that combines the optimization of a fermentation process with metabolic network design in a bilevel optimization program. The upper level maximizes space-time yield and includes mass balances of a continuous fermentation, while the lower level is based on flux balance analysis. SimulKnock predicts optimal gene deletions and finds the optimal trade-off between growth rate and product yield. Results of a case study with a genome-scale metabolic model of E. coli indicate higher space-time yields than a sequential approach using OptKnock for almost all target products considered. By leveraging SimulKnock, we reduce the gap between strain and process optimization.

Large Language Models (LLM) exhibit zero-shot mathematical reasoning capacity as a behavior emergent with scale, commonly manifesting as chain-of-thoughts (CoT) reasoning. However, multiple empirical findings suggest that this prowess is exclusive to LLMs with exorbitant sizes (beyond 50 billion parameters). Meanwhile, educational neuroscientists suggest that symbolic algebraic manipulation be introduced around the same time as arithmetic word problems to modularize language-to-formulation, symbolic manipulation of the formulation, and endgame arithmetic. In this paper, we start with the hypothesis that much smaller LMs, which are weak at multi-step reasoning, can achieve reasonable arithmetic reasoning if arithmetic word problems are posed as a formalize-then-solve task. In our architecture, which we call SYRELM, the LM serves the role of a translator to map natural language arithmetic questions into a formal language (FL) description. A symbolic solver then evaluates the FL expression to obtain the answer. A small frozen LM, equipped with an efficient low-rank adapter, is capable of generating FL expressions that incorporate natural language descriptions of the arithmetic problem (e.g., variable names and their purposes, formal expressions combining variables, etc.). We adopt policy-gradient reinforcement learning to train the adapted LM, informed by the non-differentiable symbolic solver. This marks a sharp departure from the recent development in tool-augmented LLMs, in which the external tools (e.g., calculator, Web search, etc.) are essentially detached from the learning phase of the LM. SYRELM shows massive improvements (e.g., +30.65 absolute point improvement in accuracy on the SVAMP dataset using GPT-J 6B model) over base LMs, while keeping our testbed easy to diagnose, interpret and within reach of most researchers., Comment: AAAI 2024

Large-scale deployment of fully autonomous vehicles requires a very high degree of robustness to unstructured traffic, and weather conditions, and should prevent unsafe mispredictions. While there are several datasets and benchmarks focusing on segmentation for drive scenes, they are not specifically focused on safety and robustness issues. We introduce the IDD-AW dataset, which provides 5000 pairs of high-quality images with pixel-level annotations, captured under rain, fog, low light, and snow in unstructured driving conditions. As compared to other adverse weather datasets, we provide i.) more annotated images, ii.) paired Near-Infrared (NIR) image for each frame, iii.) larger label set with a 4-level label hierarchy to capture unstructured traffic conditions. We benchmark state-of-the-art models for semantic segmentation in IDD-AW. We also propose a new metric called ''Safe mean Intersection over Union (Safe mIoU)'' for hierarchical datasets which penalizes dangerous mispredictions that are not captured in the traditional definition of mean Intersection over Union (mIoU). The results show that IDD-AW is one of the most challenging datasets to date for these tasks. The dataset and code will be available here: http://iddaw.github.io., Comment: 8 pages excluding references. Accepted in WACV 2024

Despite recent advances in quantum sciences, a quantum master equation that accurately and simply characterizes open quantum dynamics across extremely long timescales and in dispersive environments is still needed. In this study, we optimize the computation of the fourth-order time-convolutionless master equation to meet this need. Early versions of this master equation required computing a multidimensional integral, limiting its use. Our master equation accounts for simultaneous relaxation and dephasing, resulting in coefficients proportional to the system's spectral density over frequency derivative. In sub-Ohmic environments, this derivative induces infrared divergence in the master equation, invalidating the second-order Bloch-Redfield master equation findings. We analyze the approach to a ground state in a generic open quantum system and demonstrate that it is not reliably computed by the Bloch-Redfield equation alone. The optimized fourth-order equation shows that the ground-state approach is accurate to second order in bath coupling regardless of the dispersion, even though it can diverge in the fourth order at zero temperature., Comment: 28 pages, 5 figures Changes: Typos in Eq. 35 and 77 fixed. Rewording of the abstract, introduction and conclusion, in response to review. Title changed to emphasise the main result

Large Language Models (LLMs) prompted to generate chain-of-thought (CoT) exhibit impressive reasoning capabilities. Recent attempts at prompt decomposition toward solving complex, multi-step reasoning problems depend on the ability of the LLM to simultaneously decompose and solve the problem. A significant disadvantage is that foundational LLMs are typically not available for fine-tuning, making adaptation computationally prohibitive. We believe (and demonstrate) that problem decomposition and solution generation are distinct capabilites, better addressed in separate modules, than by one monolithic LLM. We introduce DaSLaM, which uses a decomposition generator to decompose complex problems into subproblems that require fewer reasoning steps. These subproblems are answered by a solver. We use a relatively small (13B parameters) LM as the decomposition generator, which we train using policy gradient optimization to interact with a solver LM (regarded as black-box) and guide it through subproblems, thereby rendering our method solver-agnostic. Evaluation on multiple different reasoning datasets reveal that with our method, a 175 billion parameter LM (text-davinci-003) can produce competitive or even better performance, compared to its orders-of-magnitude larger successor, GPT-4. Additionally, we show that DaSLaM is not limited by the solver's capabilities as a function of scale; e.g., solver LMs with diverse sizes give significant performance improvement with our solver-agnostic decomposition technique. Exhaustive ablation studies evince the superiority of our modular finetuning technique over exorbitantly large decomposer LLMs, based on prompting alone., Comment: EMNLP 2023 (Typos corrected)

Graph partitioning aims to divide a graph into disjoint subsets while optimizing a specific partitioning objective. The majority of formulations related to graph partitioning exhibit NP-hardness due to their combinatorial nature. Conventional methods, like approximation algorithms or heuristics, are designed for distinct partitioning objectives and fail to achieve generalization across other important partitioning objectives. Recently machine learning-based methods have been developed that learn directly from data. Further, these methods have a distinct advantage of utilizing node features that carry additional information. However, these methods assume differentiability of target partitioning objective functions and cannot generalize for an unknown number of partitions, i.e., they assume the number of partitions is provided in advance. In this study, we develop NeuroCUT with two key innovations over previous methodologies. First, by leveraging a reinforcement learning-based framework over node representations derived from a graph neural network and positional features, NeuroCUT can accommodate any optimization objective, even those with non-differentiable functions. Second, we decouple the parameter space and the partition count making NeuroCUT inductive to any unseen number of partition, which is provided at query time. Through empirical evaluation, we demonstrate that NeuroCUT excels in identifying high-quality partitions, showcases strong generalization across a wide spectrum of partitioning objectives, and exhibits strong generalization to unseen partition count., Comment: To appear in Knowledge Discovery and Data Mining(KDD), 2024

GNNs, like other deep learning models, are data and computation hungry. There is a pressing need to scale training of GNNs on large datasets to enable their usage on low-resource environments. Graph distillation is an effort in that direction with the aim to construct a smaller synthetic training set from the original training data without significantly compromising model performance. While initial efforts are promising, this work is motivated by two key observations: (1) Existing graph distillation algorithms themselves rely on training with the full dataset, which undermines the very premise of graph distillation. (2) The distillation process is specific to the target GNN architecture and hyper-parameters and thus not robust to changes in the modeling pipeline. We circumvent these limitations by designing a distillation algorithm called Mirage for graph classification. Mirage is built on the insight that a message-passing GNN decomposes the input graph into a multiset of computation trees. Furthermore, the frequency distribution of computation trees is often skewed in nature, enabling us to condense this data into a concise distilled summary. By compressing the computation data itself, as opposed to emulating gradient flows on the original training set-a prevalent approach to date-Mirage transforms into an unsupervised and architecture-agnostic distillation algorithm. Extensive benchmarking on real-world datasets underscores Mirage's superiority, showcasing enhanced generalization accuracy, data compression, and distillation efficiency when compared to state-of-the-art baselines., Comment: 14 pages, 14 figures

Despite outstanding performance in a variety of NLP tasks, recent studies have revealed that NLP models are vulnerable to adversarial attacks that slightly perturb the input to cause the models to misbehave. Among these attacks, adversarial word-level perturbations are well-studied and effective attack strategies. Since these attacks work in black-box settings, they do not require access to the model architecture or model parameters and thus can be detrimental to existing NLP applications. To perform an attack, the adversary queries the victim model many times to determine the most important words in an input text and to replace these words with their corresponding synonyms. In this work, we propose a lightweight and attack-agnostic defense whose main goal is to perplex the process of generating an adversarial example in these query-based black-box attacks; that is to fool the textual fooler. This defense, named AdvFooler, works by randomizing the latent representation of the input at inference time. Different from existing defenses, AdvFooler does not necessitate additional computational overhead during training nor relies on assumptions about the potential adversarial perturbation set while having a negligible impact on the model's accuracy. Our theoretical and empirical analyses highlight the significance of robustness resulting from confusing the adversary via randomizing the latent space, as well as the impact of randomization on clean accuracy. Finally, we empirically demonstrate near state-of-the-art robustness of AdvFooler against representative adversarial word-level attacks on two benchmark datasets., Comment: Accepted to Findings of ACL 2024

Deep graph generative modeling has gained enormous attraction in recent years due to its impressive ability to directly learn the underlying hidden graph distribution. Despite their initial success, these techniques, like much of the existing deep generative methods, require a large number of training samples to learn a good model. Unfortunately, large number of training samples may not always be available in scenarios such as drug discovery for rare diseases. At the same time, recent advances in few-shot learning have opened door to applications where available training data is limited. In this work, we introduce the hitherto unexplored paradigm of few-shot graph generative modeling. Towards this, we develop GSHOT, a meta-learning based framework for few-shot labeled graph generative modeling. GSHOT learns to transfer meta-knowledge from similar auxiliary graph datasets. Utilizing these prior experiences, GSHOT quickly adapts to an unseen graph dataset through self-paced fine-tuning. Through extensive experiments on datasets from diverse domains having limited training samples, we establish that GSHOT generates graphs of superior fidelity compared to existing baselines., Comment: Accepted in Learning on Graph Conference (LOG,2023),https://openreview.net/forum?id=Hy9K2WiVwW

Graph neural networks (GNNs), in general, are built on the assumption of a static set of features characterizing each node in a graph. This assumption is often violated in practice. Existing methods partly address this issue through feature imputation. However, these techniques (i) assume uniformity of feature set across nodes, (ii) are transductive by nature, and (iii) fail to work when features are added or removed over time. In this work, we address these limitations through a novel GNN framework called GRAFENNE. GRAFENNE performs a novel allotropic transformation on the original graph, wherein the nodes and features are decoupled through a bipartite encoding. Through a carefully chosen message passing framework on the allotropic transformation, we make the model parameter size independent of the number of features and thereby inductive to both unseen nodes and features. We prove that GRAFENNE is at least as expressive as any of the existing message-passing GNNs in terms of Weisfeiler-Leman tests, and therefore, the additional inductivity to unseen features does not come at the cost of expressivity. In addition, as demonstrated over four real-world graphs, GRAFENNE empowers the underlying GNN with high empirical efficacy and the ability to learn in continual fashion over streaming feature sets., Comment: 17 pages, 4 figures and 9 tables. Accepted in ICML 2023, DOI will be updated once it is available

Optimization of atomic structures presents a challenging problem, due to their highly rough and non-convex energy landscape, with wide applications in the fields of drug design, materials discovery, and mechanics. Here, we present a graph reinforcement learning approach, StriderNET, that learns a policy to displace the atoms towards low energy configurations. We evaluate the performance of StriderNET on three complex atomic systems, namely, binary Lennard-Jones particles, calcium silicate hydrates gel, and disordered silicon. We show that StriderNET outperforms all classical optimization algorithms and enables the discovery of a lower energy minimum. In addition, StriderNET exhibits a higher rate of reaching minima with energies, as confirmed by the average over multiple realizations. Finally, we show that StriderNET exhibits inductivity to unseen system sizes that are an order of magnitude different from the training system.

Recent studies have shown that velocity differences of very wide binary stars, measured to high precision with GAIA, can provide an interesting test for modified-gravity theories which emulate dark matter; in essence, MOND-like theories (with external field effect included) predict that wide binaries (wider than ~ 7 kAU) should orbit ~ 15% faster than Newtonian for similar orbits; such a shift is readily detectable in principle in the sample of 9,000 candidate systems selected from GAIA EDR3 by Pittordis and Sutherland (2022; PS22). However, the main obstacle at present is the observed "fat tail" of candidate wide-binary systems with velocity differences at ~1.5 - 6x circular velocity; this tail cannot be bound pure binary systems, but a possible explanation of the tail is triple or quadruple systems with unresolved or undetected additional star(s). While this tail can be modelled and statistically subtracted, obtaining an accurate model for the triple population is crucial for a robust test for modified gravity. In this paper we explore prospects for observationally constraining the triple population: we simulate a population of hierarchical triple systems "observed" as in PS22 at random epochs and viewing angles; for each simulated system we evaluate various possible methods for detecting the third star, including GAIA astrometry, radial velocity drift, and several imaging methods from direct Rubin images, speckle imaging and coronagraphic imaging. Results are generally encouraging, in that typically 90% of the triple systems in the key regions of parameter space are detectable; there is a moderate "dead zone" of cool brown-dwarf companions at ~ 25-100 AU separation which are not detectable with any of our baseline methods. A large but feasible observing campaign can greatly clarify the effect of the triple/quadruple population and make the gravity test decisive., Comment: Latex, 10 pages, 3 figures. Accepted by Open Journal of Astrophysics. Version 3 fixed problem with .bbl file

Knowledge graph (KG) embedding techniques use structured relationships between entities to learn low-dimensional representations of entities and relations. The traditional KG embedding techniques (such as TransE and DistMult) estimate these embeddings via simple models developed over observed KG triplets. These approaches differ in their triplet scoring loss functions. As these models only use the observed triplets to estimate the embeddings, they are prone to suffer through data sparsity that usually occurs in the real-world knowledge graphs, i.e., the lack of enough triplets per entity. To settle this issue, we propose an efficient method to augment the number of triplets to address the problem of data sparsity. We use random walks to create additional triplets, such that the relations carried by these introduced triplets entail the metapath induced by the random walks. We also provide approaches to accurately and efficiently filter out informative metapaths from the possible set of metapaths, induced by the random walks. The proposed approaches are model-agnostic, and the augmented training dataset can be used with any KG embedding approach out of the box. Experimental results obtained on the benchmark datasets show the advantages of the proposed approach.

The key to e-commerce search is how to best utilize the large yet noisy log data. In this paper, we present our embedding-based model for grocery search at Instacart. The system learns query and product representations with a two-tower transformer-based encoder architecture. To tackle the cold-start problem, we focus on content-based features. To train the model efficiently on noisy data, we propose a self-adversarial learning method and a cascade training method. AccOn an offline human evaluation dataset, we achieve 10% relative improvement in RECALL@20, and for online A/B testing, we achieve 4.1% cart-adds per search (CAPS) and 1.5% gross merchandise value (GMV) improvement. We describe how we train and deploy the embedding based search model and give a detailed analysis of the effectiveness of our method., Comment: Accepted by SIGIR eCom, July 15, 2022