Research
On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

168,742 papers · 148 categories

Trend · papers per month

199398597796 · Jun 202019922001200920172026
48 results for LLM optimization

LLMs struggle to optimize hyperparameters efficiently, but hybrid methods can improve performance.

problem Optimizing hyperparameters of small language models using LLMs.
method Comparison of classical HPO algorithms and LLM-based methods, introducing Centaur hybrid approach.
result Hybrid Centaur approach achieves best results, outperforming classical and pure LLM methods.

Diffusion LLMs can efficiently generate harmful prompts for adversarial testing.

problem Generating harmful prompts for adversarial testing is resource-intensive and costly.
method Transformed adversarial prompt optimization into an efficient inference task using pretrained Diffusion LLMs.
result Only a few conditional samples are required to generate harmful prompts with high reward.

Physics-inspired methods optimize SVD compression of LLMs.

problem Efficiently compressing large language models (LLMs) using SVD.
method FermiGrad for globally optimal rank selection and PivGa for lossless compression.
result Global optimization of SVD ranks and lossless compression of low-rank factors.

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.

The paper optimizes LLM accuracy by stopping early based on consistent answers.

problem Improving LLM accuracy in math and reasoning problems.
method Bayesian stopping policy to save on sampling costs, tracking only the L-1 most frequent answer counts.
result The L=3 stopping policy is sufficient for asymptotic optimality and significantly reduces inference costs.

New framework improves LLM performance by avoiding forgetting during sequential training stages.

problem Forgetting during sequential training stages of LLMs.
method Proposes a joint post-training framework with theoretical convergence guarantees.
result Empirically outperforms sequential post-training framework by up to 23%.

SOCRATES uses LLMs to automate simulation optimization of complex systems.

problem Optimizing complex, expensive-to-sample stochastic systems.
method Two-stage procedure: replica construction and meta-optimization.
result Adaptive hybrid optimization schedule for real systems.

Improves drug properties using a novel LLM and reinforcement learning.

problem Optimizing drug properties while retaining chemical stability.
method Structured Policy Optimization (SPO) for fine-tuning a large language model.
result Enhanced drug properties across multiple target objectives.

This paper proposes a new portfolio allocation method using LLMs to outperform traditional strategies.

problem Persistent tradeoff between risk and return in portfolio management.
method Follow-the-leader approach with sentiment-based trade filtering and LLM-driven hedging.
result Empirical results show a 69% increase in annualized returns and 119% in Sharpe ratio compared to SPY buy-and-hold.

Inference-aware meta-alignment of LLMs reduces computational cost.

problem Aligning LLMs to diverse human preferences is challenging due to conflicting criteria.
method IAMA trains a base model to be aligned to multiple tasks via different inference-time alignment algorithms, using non-linear GRPO for optimization.
result IAMA enables effective alignment of LLMs to multiple criteria with limited computational budget.

New method uses LLMs to generate detailed scientific hypotheses.

problem Generating detailed, actionable scientific hypotheses from coarse initial directions.
method Hierarchical search method that incrementally adds details to hypotheses.
result Hierarchical search method consistently outperforms strong baselines on expert-annotated hypotheses.

Develops a power-calibrated framework for LLM watermarking, optimizing tradeoffs between detectability and distortion.

problem The trade-off between detectability and semantic distortion in logit-based watermarking.
method Power-calibrated statistical framework for watermark hyperparameters, establishing explicit relationships.
result Derives practical parameter selection procedures achieving optimal tradeoffs under constraints.

PPO optimizes LLM-generated alpha weights for better trading performance.

problem Adapting LLM-generated alphas for varying market conditions.
method Proximal Policy Optimization (PPO) for dynamic alpha weight adjustment.
result PPO-optimized strategy achieves higher Sharpe ratios and smaller drawdowns.

LLMs learn to recommend models and hyperparameters from dataset metadata.

problem Model and hyperparameter selection in machine learning is challenging and resource-intensive.
method Converted datasets into metadata and prompted LLMs to recommend models and hyperparameters.
result LLMs can recommend competitive models and hyperparameters without search.

Generative AI models enhance sector-based investment portfolios, but performance varies by market conditions.

problem Improving investment performance through better stock selection in volatile markets.
method Applied LLMs from OpenAI, Google, Anthropic, DeepSeek, and xAI to select and weight stocks within S&P 500 sectors.
result LLM-weighted portfolios outperform sector indices in stable markets but underperform in volatile ones.

VERA uses variational inference to jailbreak LLMs without manual optimization.

problem Lack of principled objective for gradient-based optimization in jailbreaking LLMs.
method VERA casts black-box jailbreak prompting as a variational inference problem, training a small attacker LLM to approximate the target LLM's posterior over adversarial prompts.
result VERA achieves strong performance across various target LLMs, demonstrating the value of probabilistic inference for adversarial prompt generation.

LaTRO optimizes latent reasoning in LLMs without external reward.

problem Training LLMs to perform complex reasoning tasks.
method Formulates reasoning as latent distribution sampling and optimizes via variational approaches.
result LLMs improve reasoning and evaluation quality through self-improvement.

This paper fine-tunes LLMs for stock return prediction using financial news.

problem Improving stock return forecasting accuracy using LLMs.
method Fine-tuning LLMs with text and forecasting modules, comparing encoder-only and decoder-only models, and integrating token-level representations.
result LLMs' aggregated token-level embeddings enhance return predictions for long-only and long-short portfolios.

Study compares DSPy teleprompter algorithms for aligning LLM evaluations with human annotations.

problem Aligning LLM evaluation metrics with human annotations.
method Comparative analysis of five teleprompter algorithms within the DSPy framework.
result Certain teleprompters outperform others in detecting hallucinations.

Pion optimizes LLMs by preserving weight matrix singular values.

problem Training large language models (LLMs) with standard optimizers leads to unstable weight matrices.
method Pion uses orthogonal transformations to update weight matrices, preserving their singular values.
result Pion offers a stable alternative to standard optimizers for LLM pretraining and finetuning.

Optimization-based pruning eliminates backpropagation for large language models.

problem Suboptimal pruning performance due to heuristic metrics.
method Optimization of Bernoulli distribution to learn pruning masks without backpropagation.
result Efficient pruning of large language models with improved performance.

Best-of-\infty improves LLM performance by efficiently allocating inference-time computation.

problem Achieving optimal performance in test-time LLM ensembling with infinite budget.
method Adaptive generation scheme and weighted ensembles of LLMs, formulated as mixed-integer linear program.
result Optimal ensemble weighting improves performance over individual models.

FSPO optimizes synthetic preferences for LLM personalization.

problem Personalizing large language models for diverse users.
method FSPO reframes reward modeling as a meta-learning problem, using few labeled preferences and synthetic data.
result FSPO achieves high winrates in personalized responses, both synthetic and real.

LLMs show surprising confidence in their answers, beyond just tokens.

problem LLMs lack meaningful confidence estimates for their responses.
method Semantic calibration test based on local loss optimality and equivalence classes.
result Base LLMs are semantically calibrated across tasks, contrary to expectations.

ATLAS uses LLMs to adaptively trade by optimizing prompts and coordinating agents.

problem Adapting LLMs for real-time financial decision-making in noisy markets.
method ATLAS integrates structured market data, uses Adaptive-OPRO for prompt optimization, and employs multi-agent coordination.
result Adaptive-OPRO consistently outperforms fixed prompts in financial trading.

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.

Duel-Evolve uses LLM self-preferences for test-time optimization of discrete outputs.

problem Optimizing LLM outputs at test time with limited or unreliable scalar rewards.
method Duel-Evolve uses pairwise comparisons from the LLM to guide optimization, aggregating them via a Bayesian Bradley-Terry model.
result Achieves significant improvement over existing methods in accuracy.

New methods improve LLM preference optimization by intelligently weighting multiple reference models.

problem Improving LLM preference optimization with multiple reference models.
method Introducing four new weighting strategies for multiple-reference preference optimization.
result All four new weighting strategies outperform current methods on preference accuracy.