CARROT optimizes LLM routing by choosing the cheapest and most accurate model.
problem Optimizing query routing to the most cost-effective LLM for a given task.
method CARROT uses cost and performance estimates to select the best LLM for each query.
result CARROT is minimax optimal, selecting the best LLM for any query.
Optimizes query routing to LLMs under cost and resource constraints.
problem Non-uniform or adversarial batching in per-query routing methods leads to cost inefficiency.
method Batch-level, resource-aware routing framework that jointly optimizes model assignment for each batch.
result Robust routing framework improves accuracy by 1-14% over non-robust methods.
RACER optimizes LLM-as-judge accuracy with dynamic reasoning selection.
problem Balancing reasoning accuracy with computational cost in LLM-as-judge settings.
method Formulates routing as a constrained distributionally robust optimization problem, accounting for distribution shift via KL-divergence uncertainty set.
result RACER achieves superior accuracy-cost trade-offs under distribution shift.
Bayesian MoE framework improves LLMs' uncertainty detection.
problem Brittleness and overconfidence in deterministic routing of LLMs.
method Structured Bayesian routing in weight-space, logit-space, and selection-space.
result Significant improvements in routing stability, calibration, and OoD detection.
End-to-end framework learns LLM routing from observational data.
problem Compounding errors in decoupled approaches and reliance on full-feedback data.
method Causal end-to-end framework minimizing decision-making regret from observational data.
result Method outperforms existing baselines across different embedding models.
MM-DREX adapts LLM experts for financial trading via dynamic routing.
problem Challenges of non-stationary financial markets and static expert designs.
method MM-DREX uses a VLM-powered dynamic router to allocate expert weights and designs heterogeneous trading experts.
result Significantly outperforms 15 baselines across key metrics.
Meta-Router optimizes LLM selection using gold-standard and preference-based data.
problem Training a high-quality LLM router with combined data sources is challenging due to bias and scarcity.
method Developed an integrative causal router training framework to correct bias and improve routing accuracy.
result Our approach delivers more accurate routing and improves the trade-off between cost and quality.
The paper optimizes LLM inference systems through queueing theory.
problem Efficient LLM inference for AI agents under various routing topologies.
method Developed a fluid-limit framework for multi-class batched processing networks under K-FCFS scheduling.
result Proved that work-conserving scheduling algorithms maximize throughput for LLM inference.
PEAR dynamically reconfigures agent roles to prevent persistent biases in multi-agent debates.
problem Persistent positional biases and sensitivity to role assignments in fixed topologies.
method Dynamic reconfiguration of agent roles and sparse topologies based on evolving agent states.
result Significantly improves average accuracy over debate baselines across multiple reasoning benchmarks.
CoE modularizes LLMs for scalable, cost-effective AI systems.
problem Challenges in scalability, cost, and customization of monolithic LLMs.
method Composition of Experts (CoE) leverages multiple expert LLMs with a router for dynamic selection.
result CoE achieves superior performance with reduced computational overhead and cost.
LMoE uses LLMs to improve stock trading by selecting experts based on textual and price data.
problem Traditional neural network-based router selection in MoE models is suboptimal and ignores textual data.
method Proposes LLMoE, using LLMs as routers to select experts based on historical price data and stock news.
result LLoM outperforms state-of-the-art MoE models and other deep neural network approaches.
A system for supervising decentralized finance risks using LLMs and structured evidence.
problem Supervising decentralized finance risks
method Forecast-grounded agentic supervision system
result Developed a system that scores tickets against a regulator-aligned ground truth and false-intervention rate.
DeXposure-Claw supervises decentralized finance risks by grounding LLM decisions in evidence.
problem Weak evidence leads to over-interventions by general-purpose LLM agents in decentralized finance.
method DeXposure-Claw uses a graph time-series foundation model to forecast exposure networks, turning forecasts into alerts and constraining escalation with data-health gates.
result DeXposure-Claw reduces false alarms and improves regulator alignment in decentralized finance risk supervision.
Optimized deferral improves accuracy in imbalanced settings.
problem Imbalance in expert predictions leads to suboptimal performance in two-stage learning to defer.
method Developed novel cost-sensitive learning algorithms and margin-based loss functions tailored for expert imbalance.
result MILD algorithm shows clear improvements over baselines in image classification and LLM routing tasks.
SAGE improves memory efficiency by selectively adding, merging, or ignoring new facts.
problem Efficiently managing new facts in agentic LLMs to avoid costly write-time reasoning.
method SAGE uses a von Mises-Fisher-based density estimator to score and route candidate facts.
result SAGE achieves the best average token-F1 on LoCoMo and reduces add-phase API cost by 3.4x on GPT-4o-mini.
LLMs add value in commodity portfolio construction when information set and implementation rules are held fixed.
problem Commodity portfolio construction
method Multi-Agent LLM Framework
result LLM strategies outperform Rule Agent in Sharpe terms
Proposes a new framework for uncertainty-aware LLM post-training.
problem Heterogeneous, conflicting data in large language models.
method α-Rényi variational framework for learning distributions over post-training parameters.
result Enables training examples to be softly routed across ensemble members, promoting model specialisation and providing uncertainty estimates.
TRM improves long-horizon LLM RL by masking divergent sequences.
problem Long-horizon reinforcement learning with LLMs suffers from off-policy mismatch and approximation errors.
method Derives and applies trust region bounds to control divergence, proposing Trust Region Masking.
result First non-vacuous monotonic improvement guarantees for long-horizon LLM-RL.
TRM improves long-horizon reinforcement learning for LLMs by masking divergent sequences.
problem Long-horizon reinforcement learning for LLMs suffers from off-policy mismatch and approximation errors.
method Derives and applies trust region bounds to control divergence, proposing Trust Region Masking.
result First non-vacuous monotonic improvement guarantees for long-horizon LLM-RL.
Algorithm reduces audit costs by identifying best service configurations from biased textual evidence.
problem Designing service systems from textual evidence requires accurate selection despite biased automated scoring.
method Developed PP-LUCB algorithm combining LLM scores and selective audits to minimize costs.
result Correctly identified the best model in 40/40 trials with 90% cost reduction.
TENP prunes experts and neurons in Mixture-of-Experts models for efficient deployment.
problem Efficient deployment of large language models constrained by static parameter footprint.
method Structured Trapezoidal ExpertNeuron Pruning (TENP) identifies and retains important experts and neurons.
result DeepSeek model achieves 10% better performance on code generation tasks with 40% expert sparsity.
A new method for efficient neural network fine-tuning using queryable low-rank update atoms.
problem Rigidity of static low-rank adaptation methods when input and depth-wise computation vary.
method A shared queryable memory of low-rank update atoms, allowing dynamic and context-sensitive adaptation.
result Improves final test performance and training stability compared to standard low-rank adaptation.
Capsule networks improve performance on image classification tasks with fewer parameters.
problem Improving performance of capsule networks with fewer parameters.
method Inverted dot-product attention routing, Layer Normalization, concurrent iterative routing.
result Improves performance on benchmark datasets CIFAR-10 and CIFAR-100, and performs at-par with ResNet-18.
AgensFlow learns multi-agent coordination policies from experience.
problem Difficult coordination choices in multi-agent systems built on LLMs.
method Online policy learning from repeated trajectories, treating decisions as learnable.
result Learned routing improves coordination-heavy workflows over static wiring.
Route Choice Models predict the route choices of travelers traversing an urban area. Most of the route choice models link route characteristics of alternative routes to those chosen by the drivers. The models play an important role in prediction of traffic levels on different routes and thus assist in development of ef…
MarketSenseAI system outperforms passive benchmarks by 25.2% on S&P 500, adding value over random selection.
problem Identifying alpha in stock recommendations from multi-agent LLM systems.
method Deployed multi-agent LLM equity system generating live signals, combining four specialist agents into a synthesis agent.
result Strong-buy equal-weight portfolio on S&P 500 earns +2.18%/month, significantly outperforming passive benchmarks.
JAMPR learns to solve complex VRP with time windows.
problem Vehicle routing problems with time windows and vehicle capacities.
method Joint attention to construct multiple routes concurrently.
result JAMPR outperforms existing models on different problem sizes.
A large GPS dataset reveals that a single route often covers 60% of travel observations.
problem Limited insights from small GPS datasets on route choice behavior.
method Evaluation of path generation algorithms including link penalty, link elimination, simulation, and via-node methods.
result Modified link penalty method achieves 97% coverage, significantly higher than previous studies.
Global routing has been a historically challenging problem in electronic circuit design, where the challenge is to connect a large and arbitrary number of circuit components with wires without violating the design rules for the printed circuit boards or integrated circuits. Similar routing problems also exist in the de…
We propose and systematically evaluate three strategies for training dynamically-routed artificial neural networks: graphs of learned transformations through which different input signals may take different paths. Though some approaches have advantages over others, the resulting networks are often qualitatively similar…
A new router uses attention-based reinforcement learning to solve detailed routing problems efficiently.
problem Solving detailed routing in integrated circuits while adhering to complex design rules.
method Attention-based reinforcement learning applied to track-assignment detailed routing.
result The attention router achieves over 100x acceleration compared to a genetic router without sacrificing solution quality.
The DEBS Grand Challenge 2018 is set in the context of maritime route prediction. Vessel routes are modeled as streams of Automatic Identification System (AIS) data points selected from real-world tracking data. The challenge requires to correctly estimate the destination ports and arrival times of vessel trips, as ear…
Unified routing and arbitrage with concave continuation.
problem Combining routing and arbitrage in financial markets.
method Extending AMM trade functions to negative inputs via concave continuation.
result Unified approach unifies routing and arbitrage.
A new approach integrates inventory prediction and routing optimization for better supply chain management.
problem Optimizing efficient route selection in supply chain management with uncertain inventory demand.
method Decision-focused learning approach using neural networks to directly integrate inventory prediction and routing optimization.
result Direct integration of inventory prediction and routing optimization leads to better supply chain decisions.
Sparse routing networks with co-training prevent catastrophic forgetting in continual learning.
problem Catastrophic forgetting in neural networks trained on a sequence of tasks.
method Sparse routing networks with co-training to minimize interference between dissimilar tasks.
result Sparse routing networks with co-training outperform densely connected networks on benchmarks.
Neural LNS improves vehicle routing performance.
problem Optimizing vehicle routes with constraints.
method Integrates deep learning with large neighborhood search.
result Significantly outperforms existing methods on CVRP.
Polestar optimizes public transportation routes for efficiency and user satisfaction.
problem Difficulty in finding optimal public transportation routes due to complex networks and dynamic situations.
method Developed a Public Transportation Graph (PTG) and a route search algorithm with station binding and ranking modules.
result Demonstrated superior efficiency and user satisfaction compared to existing systems.
A recently proposed method in deep learning groups multiple neurons to capsules such that each capsule represents an object or part of an object. Routing algorithms route the output of capsules from lower-level layers to upper-level layers. In this paper, we prove that state-of-the-art routing procedures decrease the e…
Capsules are the multidimensional analogue to scalar neurons in neural networks, and because they are multidimensional, much more complex routing schemes can be used to pass information forward through the network than what can be used in traditional neural networks. This work treats capsules as collections of neurons …
Capsule networks improve at detecting changes in compositionality with routing.
problem Capsule networks struggle with detecting changes in compositionality.
method Introduced a loss function based on routing entropy to improve compositionality.
result Capsule networks with the new loss function better detect changes in compositionality.
Study finds average 2.02 bps loss in automated market maker routing.
problem Measuring sub-optimality in automated market maker routing.
method Three reproducible optimal benchmarks: SCO, FVO, G-FVO; bisection-based algorithm for optimal routing.
result Average 2.02 bps loss per trade, \$24 million total loss.
Enhanced route planning with probabilistic prediction and uncertainty sets.
problem Improving route planning reliability under uncertainty.
method CQR-GAE model integrating conformal prediction and uncertainty sets.
result Significantly outperforms baseline methods in real-world traffic scenarios.
New machine learning pipeline solves dynamic vehicle routing problems efficiently.
problem Efficiently handling same day deliveries in e-commerce logistics.
method Combination of machine learning and combinatorial optimization.
result Ranked first in the EURO Meets NeurIPS Vehicle Routing Competition.
New algorithm for traffic routing in congested conditions.
problem Optimal routing in congested traffic networks.
method Congested Bandits model, UCB algorithm, iterative least squares planner.
result No-regret algorithms for congestion-aware routing.
Optimal crypto asset routing with CFMMs, including fixed costs.
problem Optimizing order execution on a network of CFMMs with fixed costs.
method Convex optimization for no fixed costs, mixed-integer convex for fixed costs, heuristics for approximate solutions.
result Approximate solutions to optimal routing and arbitrage certification problems.
DPDP combines neural heuristics with DP for vehicle routing problems.
problem Vehicle routing problems with large scale.
method Deep Policy Dynamic Programming (DPDP) that uses a neural network policy to prioritize and restrict the DP state space.
result DPDP improves upon classical DP algorithms and outperforms neural approaches for TSP, VRP, and TSPTW.
Framework improves ETF volatility forecasting by adapting to market conditions.
problem Challenges in volatility forecasting due to shifting market conditions and varying model performance.
method Risk-sensitive specialist routing using online risk-sensitive evaluation and state-dependent gating.
result Reduces forecast loss by 24% and underprediction loss by 22% compared to rolling-best baseline.
High-level synthesis (HLS) shortens the development time of hardware designs and enables faster design space exploration at a higher abstraction level. Optimization of complex applications in HLS is challenging due to the effects of implementation issues such as routing congestion. Routing congestion estimation is abse…