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.

168,695 papers · 148 categories

Trend · papers per month

count % of corpus

10 results for “LLaMA-3-8B”

Instruction Finetuning LLaMA-3-8B Model Using LoRA for Financial Named Entity Recognitionq-fin.CP

Paper fine-tunes LLaMA-3-8B for financial NER using instruction and LoRA.

problem LLMs struggle with financial NER, especially differentiating entities and amounts.

method Instruction fine-tuning combined with LoRA for parameter-efficient learning.

result Micro-F1 score of 0.894 on financial NER tasks, outperforming other models.

financial NER LLaMA-3-8B instruction fine-tuning LoRA state-of-the-art

2026-01-15abs ↗pdf ↗

NeuZip: Memory-Efficient Training and Inference with Dynamic Compression of Neural Networkscs.LG

NeuZip compresses neural network weights to save memory during training and inference.

problem Memory constraints in neural network training and inference.

method Entropy-based dynamic weight compression.

result Significant reduction in memory usage without performance loss.

neural networks memory efficiency compression floating-point numbers quantization

2024-10-28abs ↗pdf ↗

Beyond Pairwise: Empowering LLM Alignment With Ranked Choice Modelingcs.LG

RCPO uses ranked choice modeling for better LLM alignment.

problem Pairwise preference optimization limits LLM alignment.

method Unified framework combining preference optimization and ranked choice modeling.

result RCPO outperforms competitive baselines in LLM alignment.

LLM alignment Preference optimization Ranked choice modeling Maximum likelihood estimation Utility-based models

2025-10-24abs ↗pdf ↗

Learning to Rank Chain-of-Thought: Using a Small Modelcs.LG

EORM boosts LLM accuracy with a lightweight, energy-based verifier.

problem Efficiently verifying mathematical reasoning in large language models.

method Energy-based framework to rank Chain-of-Thought solutions using simple outcome labels.

result EORM boosts LLM accuracy to 90.7% on GSM8k and 63.7% on MATH with only 55M parameters.

mathematical reasoning large language models efficiency energy-based Chain-of-Thought

2025-05-21abs ↗pdf ↗

LOLA: LLM-Assisted Online Learning Algorithm for Content Experimentscs.LG

LOLA uses LLMs to optimize content delivery, outperforming traditional methods.

problem Identifying the most engaging headlines for user engagement.

method LOLA integrates LLMs with adaptive experimentation to optimize content delivery.

result LOLA outperforms traditional methods in optimizing user engagement.

content optimization user engagement LLM-assisted adaptive experimentation headline prediction

2024-06-03abs ↗pdf ↗

Harnessing Earnings Reports for Stock Predictions: A QLoRA-Enhanced LLM Approachq-fin.CP

Advanced AI model predicts stock movements post earnings reports.

problem Inaccurate stock predictions from earnings reports.

method Fine-tuned LLMs with QLoRA compression, integrating financial and market data.

result Significantly improved predictive accuracy compared to benchmarks.

stock prediction earnings reports LLMs QLoRA financial data

2024-08-13abs ↗pdf ↗

Provably Efficient Online RLHF with One-Pass Reward Modelingcs.LG

New method for RLHF reduces costs by integrating new data in one pass.

problem Continuous integration and re-optimization of models in RLHF leads to high computational and storage costs.

method Proposes a one-pass reward modeling method using online mirror descent with a tailored local norm.

result Achieves constant-time updates per iteration, enhancing both statistical and computational efficiency.

Reinforcement Learning Human Feedback Online RLHF Reward Modeling Contextual Preference Bandit

2025-02-11abs ↗pdf ↗

Unintentional Unalignment: Likelihood Displacement in Direct Preference Optimizationcs.LG

Training models to prefer certain responses can unintentionally shift probability to harmful ones.

problem Likelihood displacement in DPO models, leading to unintended unalignment.

method Characterized and mitigated likelihood displacement using CHES score.

result Training models to prefer certain responses can unintentionally shift probability mass to harmful responses.

Likelihood Displacement Direct Preference Optimization Unintentional Unalignment CHES Score Data Curation

2024-10-11abs ↗pdf ↗

LoLCATs: On Low-Rank Linearizing of Large Language Modelscs.LG

LoLCATs improves linearized LLM quality with less memory and compute.

problem Linearizing large language models (LLMs) often degrades model quality and requires expensive training.

method Two-step method: attention transfer and low-rank adaptation.

result Significant improvement in linearizing quality with 20+ points on 5-shot MMLU.

language models linearization attention transfer low-rank adaptation large language models

2024-10-14abs ↗pdf ↗

Self-Play Preference Optimization for Language Model Alignmentcs.LG

SPPO optimizes language model alignment by treating preferences as a game and achieving state-of-the-art performance.

problem Capturing intransitivity and irrationality in human preferences for accurate language model alignment.

method Self-play-based approach to identify Nash equilibrium policy through iterative policy updates.

result SPPO achieves state-of-the-art win-rate of 28.53% on AlpacaEval 2.0 without external supervision.

language model alignment reinforcement learning human preferences Nash equilibrium preference optimization

2024-05-01abs ↗pdf ↗