prompt stringlengths 3.53k 14.6k | response stringlengths 722 9.13k |
|---|---|
You are a research assistant specializing in computer algorithms.
Your task is to apply a mechanism-level method delta to structured details of an existing algorithm and produce structured details of the improved algorithm.
You must:
1. Faithfully apply the given method delta.
2. Preserve the research problem context u... | {"problem_definition": "Reconstruct an anisotropic reflectivity image of a scene from multistatic travel-time measurements of electromagnetic or acoustic signals, formulated as a tensor tomography inverse problem.", "method_name": "Truncated singular value expansion (SVE) inversion of the normal Radon transform", "meth... |
You are a research assistant specializing in computer algorithms.
Your task is to apply a mechanism-level method delta to structured details of an existing algorithm and produce structured details of the improved algorithm.
You must:
1. Faithfully apply the given method delta.
2. Preserve the research problem context u... | {"problem_definition": "Minimize cumulative KL-regularized regret in a stochastic K-armed bandit, where the per-round objective is J(π)=E_{a∼π}[r(a)] - η^{-1}KL(π∥π_ref) and regret is measured against the optimal policy π* for this regularized objective.", "method_name": "KL-UCB (variant for multi-armed bandits)", "met... |
You are a research assistant specializing in computer algorithms.
Your task is to apply a mechanism-level method delta to structured details of an existing algorithm and produce structured details of the improved algorithm.
You must:
1. Faithfully apply the given method delta.
2. Preserve the research problem context u... | {"problem_definition": "Verifying whether an output produced by a large language model follows the instruction used to prompt it, in a universal and general-purpose way, without assuming instruction type or LLM.", "method_name": "NSVIF (Neuro-Symbolic Verification on Instruction Following of LLMs)", "method_type": "neu... |
You are a research assistant specializing in computer algorithms.
Your task is to apply a mechanism-level method delta to structured details of an existing algorithm and produce structured details of the improved algorithm.
You must:
1. Faithfully apply the given method delta.
2. Preserve the research problem context u... | {"problem_definition": "The paper addresses the Traveling Salesman Problem (TSP) in 2D Euclidean space, an NP-hard combinatorial optimization problem, with the goal of learning to produce high-quality tours from pre-existing heuristic solution datasets, ultimately surpassing the original heuristics.", "method_name": "O... |
You are a research assistant specializing in computer algorithms.
Your task is to apply a mechanism-level method delta to structured details of an existing algorithm and produce structured details of the improved algorithm.
You must:
1. Faithfully apply the given method delta.
2. Preserve the research problem context u... | {"problem_definition": "The green bin packing (GBP) problem: given items with sizes in (0,1], parameters β≥0 and G∈[0,1], and unit-capacity bins, pack all items to minimize the number of bins used plus total energy consumption, where a bin of load x consumes energy max{0, β(x−G)}. The constrained variant (CGBP) minimiz... |
You are a research assistant specializing in computer algorithms.
Your task is to apply a mechanism-level method delta to structured details of an existing algorithm and produce structured details of the improved algorithm.
You must:
1. Faithfully apply the given method delta.
2. Preserve the research problem context u... | {"problem_definition": "Plan speed-admissible trajectories for multiple agents starting and ending at a depot to intercept a set of moving targets with time windows, satisfying capacity constraints, minimizing total travel distance.", "method_name": "Branch-and-Price with Relaxed Continuity (BPRC)", "method_type": "Exa... |
You are a research assistant specializing in computer algorithms.
Your task is to apply a mechanism-level method delta to structured details of an existing algorithm and produce structured details of the improved algorithm.
You must:
1. Faithfully apply the given method delta.
2. Preserve the research problem context u... | {"problem_definition": "Time-varying multi-agent coverage control: deploy mobile agents to optimally track a continuously evolving target density by minimizing the instantaneous Wasserstein distance between the agents' empirical measure and the target measure.", "method_name": "Time-Varying Optimal Transport Coverage C... |
You are a research assistant specializing in computer algorithms.
Your task is to apply a mechanism-level method delta to structured details of an existing algorithm and produce structured details of the improved algorithm.
You must:
1. Faithfully apply the given method delta.
2. Preserve the research problem context u... | {"problem_definition": "Optimize edge directions and weights in a mixed guidance graph for Lifelong Multi-Agent Path Finding (LMAPF) to maximize throughput by providing strict directional guidance that reduces head-on collisions and traffic congestion.", "method_name": "Mixed Guidance Graph Optimization (MGGO)", "metho... |
You are a research assistant specializing in computer algorithms.
Your task is to apply a mechanism-level method delta to structured details of an existing algorithm and produce structured details of the improved algorithm.
You must:
1. Faithfully apply the given method delta.
2. Preserve the research problem context u... | {"problem_definition": "LLM-based production systems suffer from stochastic judgment variance, poor long-context utilization with critical information loss in the middle of prompts, and unpredictable cost/latency, making governance, auditability, and consistent control difficult. PARCER addresses this by imposing a dec... |
You are a research assistant specializing in computer algorithms.
Your task is to apply a mechanism-level method delta to structured details of an existing algorithm and produce structured details of the improved algorithm.
You must:
1. Faithfully apply the given method delta.
2. Preserve the research problem context u... | {"problem_definition": "Collision-free multi-agent path finding (MAPF) in partially observable grids, minimizing total travel time or number of steps for all agents while avoiding vertex and edge conflicts.", "method_name": "PC 2 P", "method_type": "multi-agent reinforcement learning (Q-learning-based MARL)", "core_ide... |
You are a research assistant specializing in computer algorithms.
Your task is to apply a mechanism-level method delta to structured details of an existing algorithm and produce structured details of the improved algorithm.
You must:
1. Faithfully apply the given method delta.
2. Preserve the research problem context u... | {"problem_definition": "Enhancing speech-to-speech translation (S2ST) capability of large language models (LLMs) under the constraint of scarce paired S2ST data.", "method_name": "PROST-LLM (PROgressive Speech-to-speech Translation)", "method_type": "LLM fine-tuning framework", "core_idea": "Progressively enhance LLM S... |
You are a research assistant specializing in computer algorithms.
Your task is to apply a mechanism-level method delta to structured details of an existing algorithm and produce structured details of the improved algorithm.
You must:
1. Faithfully apply the given method delta.
2. Preserve the research problem context u... | {"problem_definition": "Multi-Agent Path Finding (MAPF): compute collision-free paths for a team of agents navigating to their respective goals on a grid, minimizing sum-of-costs. The paper addresses the representational bottleneck of pairwise GNNs in dense, highly-coupled multi-agent coordination by proposing hypergra... |
You are a research assistant specializing in computer algorithms.
Your task is to apply a mechanism-level method delta to structured details of an existing algorithm and produce structured details of the improved algorithm.
You must:
1. Faithfully apply the given method delta.
2. Preserve the research problem context u... | {"problem_definition": "Automated heuristic design for combinatorial optimization problems using LLMs, where existing methods rely on fixed evolutionary rules and static prompt templates, leading to myopic generation, redundant evaluations, and limited reasoning about heuristic derivation.", "method_name": "PathWise (P... |
You are a research assistant specializing in computer algorithms.
Your task is to apply a mechanism-level method delta to structured details of an existing algorithm and produce structured details of the improved algorithm.
You must:
1. Faithfully apply the given method delta.
2. Preserve the research problem context u... | {"problem_definition": "Path planning in grid environments: find an optimal and collision-free path from a start node to a terminal node.", "method_name": "Pheromone-Focused Ant Colony Optimization (PFACO)", "method_type": "heuristic optimization", "core_idea": "PFACO enhances ACO by concentrating initial pheromone on ... |
You are a research assistant specializing in computer algorithms.
Your task is to apply a mechanism-level method delta to structured details of an existing algorithm and produce structured details of the improved algorithm.
You must:
1. Faithfully apply the given method delta.
2. Preserve the research problem context u... | {"problem_definition": "Classical planning seeks sequences of actions from an initial state to a goal state; the paper addresses the problem of deriving efficient heuristics to guide classical planning search by reformulating planning heuristics as estimates of goal intentionality from a goal recognition perspective.",... |
You are a research assistant specializing in computer algorithms.
Your task is to apply a mechanism-level method delta to structured details of an existing algorithm and produce structured details of the improved algorithm.
You must:
1. Faithfully apply the given method delta.
2. Preserve the research problem context u... | {"problem_definition": "Evaluating GNN expressiveness in capturing fundamental graph properties is an open challenge; existing benchmarks and frameworks lack systematic property-driven evaluation, and the impact of global pooling on expressiveness is unexplored.", "method_name": "Property-Driven GNN Expressiveness Eval... |
You are a research assistant specializing in computer algorithms.
Your task is to apply a mechanism-level method delta to structured details of an existing algorithm and produce structured details of the improved algorithm.
You must:
1. Faithfully apply the given method delta.
2. Preserve the research problem context u... | {"problem_definition": "Estimate the spectral gap (or singular gap) of an ergodic Markov chain on a discrete state space, which serves as a proxy for the mixing time / relaxation time, using a quantum algorithm with quasi-optimal query complexity.", "method_name": "Quasi-optimal quantum Markov chain spectral gap estima... |
You are a research assistant specializing in computer algorithms.
Your task is to apply a mechanism-level method delta to structured details of an existing algorithm and produce structured details of the improved algorithm.
You must:
1. Faithfully apply the given method delta.
2. Preserve the research problem context u... | {"problem_definition": "Improving empathetic response generation in speech-based LLMs by capturing emotion nuances from speech, particularly enhancing emotional exploration (question-asking) alongside emotional reaction, which are overlooked by text-only and standard speech-LLM approaches.", "method_name": "RE-LLM (Ref... |
You are a research assistant specializing in computer algorithms.
Your task is to apply a mechanism-level method delta to structured details of an existing algorithm and produce structured details of the improved algorithm.
You must:
1. Faithfully apply the given method delta.
2. Preserve the research problem context u... | {"problem_definition": "Minimize policy regret in Rising Multi-Armed Bandits (RMAB) with a known finite horizon, where expected rewards increase with the number of plays, by selecting arms that maximize cumulative reward over the remaining horizon.", "method_name": "CURE-UCB (CUmulative Reward Estimation UCB)", "method... |
You are a research assistant specializing in computer algorithms.
Your task is to apply a mechanism-level method delta to structured details of an existing algorithm and produce structured details of the improved algorithm.
You must:
1. Faithfully apply the given method delta.
2. Preserve the research problem context u... | {"problem_definition": "The Bandwidth Coloring Problem (BCP) generalizes graph coloring by requiring that adjacent vertices receive colors separated by at least a given positive integer edge weight d(u,v). The goal is to find a coloring c: V -> N minimizing the span, i.e., the maximum color used, subject to |c(u)-c(v)|... |
You are a research assistant specializing in computer algorithms.
Your task is to apply a mechanism-level method delta to structured details of an existing algorithm and produce structured details of the improved algorithm.
You must:
1. Faithfully apply the given method delta.
2. Preserve the research problem context u... | {"problem_definition": "Graph Neural Networks (GNNs) suffer from reduced generalization due to spurious correlations between irrelevant node features/representations and labels, which occur under both IID and OOD settings; the task is to identify and mitigate such spurious correlations to improve node classification ge... |
You are a research assistant specializing in computer algorithms.
Your task is to apply a mechanism-level method delta to structured details of an existing algorithm and produce structured details of the improved algorithm.
You must:
1. Faithfully apply the given method delta.
2. Preserve the research problem context u... | {"problem_definition": "Optimization of batch acquisition functions (qEHVI) in multi-objective Bayesian optimization, particularly over finite discrete candidate sets, where gradient-based continuous optimizers such as SLSQP suffer from local optima, vanishing gradients, and violations of discrete constraints.", "metho... |
You are a research assistant specializing in computer algorithms.
Your task is to apply a mechanism-level method delta to structured details of an existing algorithm and produce structured details of the improved algorithm.
You must:
1. Faithfully apply the given method delta.
2. Preserve the research problem context u... | {"problem_definition": "Traditional deterministic heuristic tour constructors for the Traveling Salesman Problem (e.g., Farthest/Nearest Insertion) are computationally efficient but produce only a single tour per instance, limiting exploration and often getting stuck in local optima. Neural-based heuristic constructors... |
You are a research assistant specializing in computer algorithms.
Your task is to apply a mechanism-level method delta to structured details of an existing algorithm and produce structured details of the improved algorithm.
You must:
1. Faithfully apply the given method delta.
2. Preserve the research problem context u... | {"problem_definition": "Stochastic multi-armed bandit with limited control variates (MAB-LCV): in each round, the learner selects an arm and observes its reward, but the associated control variate (side information correlated with reward) is only available with some probability. The goal is to minimize cumulative regre... |
You are a research assistant specializing in computer algorithms.
Your task is to apply a mechanism-level method delta to structured details of an existing algorithm and produce structured details of the improved algorithm.
You must:
1. Faithfully apply the given method delta.
2. Preserve the research problem context u... | {"problem_definition": "Investigate whether expert routing in sparse Mixture-of-Experts transformers exhibits task-conditioned structure by introducing routing signatures to summarize expert activation patterns and testing their separability by task category.", "method_name": "Routing Signatures (implemented in MOE-XRA... |
You are a research assistant specializing in computer algorithms.
Your task is to apply a mechanism-level method delta to structured details of an existing algorithm and produce structured details of the improved algorithm.
You must:
1. Faithfully apply the given method delta.
2. Preserve the research problem context u... | {"problem_definition": "Whether end-to-end speech LLMs behave like ASR→LLM cascades on tasks where the transcript contains sufficient information for the task label, and how to evaluate this behavior while controlling for the reasoning capabilities of the LLM backbone.", "method_name": "Matched-backbone behavioral test... |
You are a research assistant specializing in computer algorithms.
Your task is to apply a mechanism-level method delta to structured details of an existing algorithm and produce structured details of the improved algorithm.
You must:
1. Faithfully apply the given method delta.
2. Preserve the research problem context u... | {"problem_definition": "The Colored Knapsack Problem (CKP): given items with profit, weight, and color, select a subset of maximum total profit that fits in a knapsack of capacity b and can be ordered so that no two consecutive items share the same color.", "method_name": "DP1 and DP2", "method_type": "dynamic programm... |
You are a research assistant specializing in computer algorithms.
Your task is to apply a mechanism-level method delta to structured details of an existing algorithm and produce structured details of the improved algorithm.
You must:
1. Faithfully apply the given method delta.
2. Preserve the research problem context u... | {"problem_definition": "Generate large-scale, reliable, and diverse urban mobility trajectories from contextual inputs such as origin and destination, while avoiding reliance on real private trajectory data.", "method_name": "TrajGPT-R", "method_type": "generative model (Transformer with reinforcement learning fine-tun... |
You are a research assistant specializing in computer algorithms.
Your task is to apply a mechanism-level method delta to structured details of an existing algorithm and produce structured details of the improved algorithm.
You must:
1. Faithfully apply the given method delta.
2. Preserve the research problem context u... | {"problem_definition": "Selecting among multiple strategic LLM providers for user queries where providers bid costs and the user must learn which provider performs best for each query context, while eliciting truthful cost reports and maximizing user utility.", "method_name": "TRCM-UCB OPT", "method_type": "contextual ... |
You are a research assistant specializing in computer algorithms.
Your task is to apply a mechanism-level method delta to structured details of an existing algorithm and produce structured details of the improved algorithm.
You must:
1. Faithfully apply the given method delta.
2. Preserve the research problem context u... | {"problem_definition": "Adapting and integrating Large Language Models (LLMs) to generate and optimize proprietary programming languages used in Industrial Process Automation (IPA), such as PLC and RAPID, despite data scarcity, domain specificity, and heterogeneous data modalities.", "method_name": "LLM-based Industria... |
You are a research assistant specializing in computer algorithms.
Your task is to apply a mechanism-level method delta to structured details of an existing algorithm and produce structured details of the improved algorithm.
You must:
1. Faithfully apply the given method delta.
2. Preserve the research problem context u... | {"problem_definition": "Improve sample efficiency in reinforcement learning by directing exploration via optimistic value estimation, providing first-visit optimism and deep exploration, while being easy to use and compatible with any base RL algorithm.", "method_name": "Value Bonuses using Ensemble Errors (VBE)", "met... |
You are a research assistant specializing in computer algorithms.
Your task is to apply a mechanism-level method delta to structured details of an existing algorithm and produce structured details of the improved algorithm.
You must:
1. Faithfully apply the given method delta.
2. Preserve the research problem context u... | {"problem_definition": "Vehicle Routing Problem with Finite Time Horizon (VRP-FTH): a single vehicle starts at a depot, serves a subset of customers, and must return to the depot within a finite time horizon U; the objective is to maximize the number of customer requests served, with customer requests being determinist... |
You are a research assistant specializing in computer algorithms.
Your task is to apply a mechanism-level method delta to structured details of an existing algorithm and produce structured details of the improved algorithm.
You must:
1. Faithfully apply the given method delta.
2. Preserve the research problem context u... | {"problem_definition": "Protect the copyright of LLM agent trajectory datasets by watermarking them so that models trained on the watermarked data can be reliably detected in a black-box setting, despite small dataset sizes and common watermark-removal attacks such as paraphrasing and summarization.", "method_name": "A... |
You are a research assistant specializing in computer algorithms.
Your task is to apply a mechanism-level method delta to structured details of an existing algorithm and produce structured details of the improved algorithm.
You must:
1. Faithfully apply the given method delta.
2. Preserve the research problem context u... | {"problem_definition": "Investigate how large language models (LLMs) resolve knowledge conflicts between contexts attributed to different sources, and mitigate repetition bias that flips source preferences.", "method_name": "Teacher-Student Knowledge Distillation for Repetition Bias Mitigation", "method_type": "evaluat... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.