Fragmentation strictly raises optimal finite-context log-loss on Markov sources while tokenization can make a short token window equivalent to a longer source window under reliability and compression conditions.
super hub Canonical reference
LLaMA: Open and Efficient Foundation Language Models
Canonical reference. 88% of citing Pith papers cite this work as background.
471
Pith papers citing it
Background
88% of classified citations
abstract
We introduce LLaMA, a collection of foundation language models ranging from 7B to 65B parameters. We train our models on trillions of tokens, and show that it is possible to train state-of-the-art models using publicly available datasets exclusively, without resorting to proprietary and inaccessible datasets. In particular, LLaMA-13B outperforms GPT-3 (175B) on most benchmarks, and LLaMA-65B is competitive with the best models, Chinchilla-70B and PaLM-540B. We release all our models to the research community.
hub tools
citation-role summary
background 7
method 1
citation-polarity summary
claims ledger
- abstract We introduce LLaMA, a collection of foundation language models ranging from 7B to 65B parameters. We train our models on trillions of tokens, and show that it is possible to train state-of-the-art models using publicly available datasets exclusively, without resorting to proprietary and inaccessible datasets. In particular, LLaMA-13B outperforms GPT-3 (175B) on most benchmarks, and LLaMA-65B is competitive with the best models, Chinchilla-70B and PaLM-540B. We release all our models to the research community.
co-cited works
representative citing papers
Allowing each quantization group to select among multiple 4-bit grids improves accuracy over single-grid FP4 for both post-training and pre-training of LLMs.
Adaptive scheduling of interventions in discrete diffusion language models, timed to attribute-specific commitment schedules discovered with sparse autoencoders, delivers precise multi-attribute steering up to 93% strength while preserving generation quality.
SignSGD provably beats SGD by a factor of d under sparse noise via matched ℓ1-norm upper and lower bounds, with an equivalent result for Muon on matrices, and this predicts faster GPT-2 pretraining.
An adversary controlling an intermediate pipeline stage in decentralized LLM post-training can inject a backdoor that reduces alignment from 80% to 6%, with the backdoor persisting in 60% of cases even after subsequent safety training.
AgentDojo introduces an extensible evaluation framework populated with realistic agent tasks and security test cases to measure prompt injection robustness in tool-using LLM agents.
Mamba is a linear-time sequence model using input-dependent selective SSMs that achieves SOTA results across modalities and matches twice-larger Transformers on language modeling with 5x higher inference throughput.
Tree of Thoughts enables language models to solve complex planning tasks by generating, evaluating, and searching over coherent intermediate thoughts in a tree, raising Game of 24 success from 4% to 74% with GPT-4.
GPT-4-generated instruction data produces superior zero-shot performance in finetuned LLaMA models versus prior state-of-the-art data.
A proposer-solver agent pair achieves supervised-level video temporal grounding and fine-grained captioning from 2.5K unlabeled videos via self-reinforcing evolution.
BlockVLA accelerates autoregressive VLA models by 3.3x using block diffusion finetuning, with faster training convergence and better early performance on long-horizon robotic tasks.
A parallel multi-turn medical dialogue dataset spanning English and nine Indic languages is created from synthetic consultations to enable personalized AI healthcare interactions.
LiteLVLM prunes visual tokens for pixel grounding by reversing CLIP visual-text similarity to retain referent region tokens, outperforming prior methods by over 5% with 22% speedup and 2.3x memory reduction without any training.
Probabilistic circuits have an output bottleneck with convex probability combinations and a context bottleneck limited to fixed vtree-aligned partitions, making them less expressive than transformers for language data with heterogeneous dependencies, though decomposable PCs are strictly more capable
MindVLA-U1 introduces a unified streaming VLA with shared backbone, framewise memory, and language-guided action diffusion that surpasses human drivers on WOD-E2E planning metrics.
Large language models achieve macro F1 scores above 0.85 on binary nominal-versus-danger classification from CTAF radio transcripts and METAR weather data using a new synthetic dataset with a 12-category hazard taxonomy.
Gender bias and factual gender knowledge are severely entangled in language model circuits and neurons, making neuron ablation an unreliable method for debiasing.
MoLA infers a mixture of latent actions from generated future videos via modality-aware inverse dynamics models to improve robot manipulation policies.
Pretrained LLMs adapted via convolutional projections and LoRA act as efficient frozen backbones for sensor-based human activity recognition, delivering strong data efficiency and cross-dataset transfer.
DistractMIA performs output-only black-box membership inference on vision-language models by inserting semantic distractors and measuring shifts in generated text responses.
StepCodeReasoner aligns code reasoning with verifiable stepwise execution traces via print anchors and bi-level GRPO reinforcement learning, reaching SOTA results on CRUXEval (91.1%) and LiveCodeBench (86.5%) for a 7B model.
Anchor-guided variance-aware reward modeling uses two response-level anchors to resolve non-identifiability in Gaussian models of pluralistic preferences, yielding provable identification, a joint training objective, and improved RLHF performance.
Temporarily reducing the learning rate on upper-layer query and key projections during early GPT pretraining prevents premature attention specialization and improves model performance.
V-ABS is an action-observer beam search method with entropy-based adaptive weighting and an 80k-sample SFT dataset that delivers 19.7% average gains on visual reasoning tasks for MLLMs.
citing papers explorer
-
Effective Context in Transformers: An Analysis of Fragmentation and Tokenization
Fragmentation strictly raises optimal finite-context log-loss on Markov sources while tokenization can make a short token window equivalent to a longer source window under reliability and compression conditions.
-
Grid Games: The Power of Multiple Grids for Quantizing Large Language Models
Allowing each quantization group to select among multiple 4-bit grids improves accuracy over single-grid FP4 for both post-training and pre-training of LLMs.
-
Steering Without Breaking: Mechanistically Informed Interventions for Discrete Diffusion Language Models
Adaptive scheduling of interventions in discrete diffusion language models, timed to attribute-specific commitment schedules discovered with sparse autoencoders, delivers precise multi-attribute steering up to 93% strength while preserving generation quality.
-
When and Why SignSGD Outperforms SGD: A Theoretical Study Based on $\ell_1$-norm Lower Bounds
SignSGD provably beats SGD by a factor of d under sparse noise via matched ℓ1-norm upper and lower bounds, with an equivalent result for Muon on matrices, and this predicts faster GPT-2 pretraining.
-
Mamba: Linear-Time Sequence Modeling with Selective State Spaces
Mamba is a linear-time sequence model using input-dependent selective SSMs that achieves SOTA results across modalities and matches twice-larger Transformers on language modeling with 5x higher inference throughput.
-
The Expressivity Boundary of Probabilistic Circuits: A Comparison with Large Language Models
Probabilistic circuits have an output bottleneck with convex probability combinations and a context bottleneck limited to fixed vtree-aligned partitions, making them less expressive than transformers for language data with heterogeneous dependencies, though decomposable PCs are strictly more capable
-
Efficient and Adaptive Human Activity Recognition via LLM Backbones
Pretrained LLMs adapted via convolutional projections and LoRA act as efficient frozen backbones for sensor-based human activity recognition, delivering strong data efficiency and cross-dataset transfer.
-
VORT: Adaptive Power-Law Memory for NLP Transformers
VORT assigns learnable fractional orders to tokens and approximates their power-law retention kernels via sum-of-exponentials for efficient long-range dependency modeling in transformers.
-
Transformers Can Implement Preconditioned Richardson Iteration for In-Context Gaussian Kernel Regression
Standard softmax-attention transformers can approximate the Gaussian kernel ridge regression predictor by implementing preconditioned Richardson iteration during their forward pass.
-
Rollback-Free Stable Brick Structures Generation
Reinforcement learning internalizes physical stability rules for brick structures, enabling the first rollback-free generation with orders-of-magnitude faster inference.
-
Transformers Efficiently Perform In-Context Logistic Regression via Normalized Gradient Descent
Multi-layer transformers can implement in-context logistic regression by performing normalized gradient descent steps layer by layer, obtained via supervised training of a single attention layer followed by recurrent application with convergence and OOD guarantees.
-
When Graph Language Models Go Beyond Memorization
Large-scale graph language models acquire structural regularities beyond memorization, with subgraph rank correlations persisting after bootstrap and novel-subset controls, especially for high-frequency patterns.
-
Adaptive Selection of LoRA Components in Privacy-Preserving Federated Learning
AS-LoRA adaptively chooses which LoRA factor to update per layer and round using a curvature-aware second-order score, eliminating reconstruction error floors and improving performance in DP federated learning.
-
Delta-Based Neural Architecture Search: LLM Fine-Tuning via Code Diffs
Fine-tuned 7B LLMs generating unified diffs for neural architecture refinement achieve 66-75% valid rates and 64-66% mean first-epoch accuracy, outperforming full-generation baselines by large margins while cutting output length by 75-85%.
-
Echo-LoRA: Parameter-Efficient Fine-Tuning via Cross-Layer Representation Injection
Echo-LoRA raises average performance on eight commonsense reasoning benchmarks by 3.0 to 5.7 points over standard LoRA by using a training-only cross-layer echo representation that is discarded after training.
-
Low Rank Adaptation for Adversarial Perturbation
Adversarial perturbations possess an inherently low-rank structure that enables more efficient and effective black-box adversarial attacks via subspace projection.
-
Adaptive and Fine-grained Module-wise Expert Pruning for Efficient LoRA-MoE Fine-Tuning
DMEP prunes experts module-by-module in LoRA-MoE and removes load balancing after pruning, cutting trainable parameters 35-43% and raising throughput ~10% while matching or exceeding uniform baselines on reasoning tasks.
-
Incompressible Knowledge Probes: Estimating Black-Box LLM Parameter Counts via Factual Capacity
Incompressible Knowledge Probes enable log-linear estimation of LLM parameter counts from factual accuracy on obscure questions, showing continued scaling of knowledge capacity across open and closed models.
-
Can an MLP Absorb Its Own Skip Connection?
Skip-connected MLPs and residual-free MLPs of equal width represent generically disjoint function classes for common activations, with explicit impossibility proofs and a non-generic absorption condition for ReLU and GELU.
-
Learning Posterior Predictive Distributions for Node Classification from Synthetic Graph Priors
NodePFN pre-trains on synthetic graphs with controllable homophily and causal feature-label models to achieve 71.27 average accuracy on 23 node classification benchmarks without graph-specific training.
-
Local Linearity of LLMs Enables Activation Steering via Model-Based Linear Optimal Control
Local linearity of LLM layers enables LQR-based closed-loop activation steering with theoretical tracking guarantees.
-
Self-Improving Tabular Language Models via Iterative Group Alignment
TabGRAA enables self-improving tabular language models through iterative group-relative advantage alignment using modular automated quality signals like distinguishability classifiers.
-
Faster by Design: Interactive Aerodynamics via Neural Surrogates Trained on Expert-Validated CFD
A graph-based neural operator trained on expert-validated race-car CFD data reaches accuracy levels usable for early-stage interactive aerodynamic design exploration.
-
Does RL Expand the Capability Boundary of LLM Agents? A PASS@(k,T) Analysis
RL expands the capability boundary of LLM agents on compositional tool-use tasks, shown by non-converging pass curves at large k with increasing T, while SFT regresses it and the effect is absent on simpler tasks.
-
Reinforcement Learning via Value Gradient Flow
VGF solves behavior-regularized RL by transporting particles from a reference distribution to the value-induced optimal policy via discrete value-guided gradient flow.
-
On the Invariants of Softmax Attention
Softmax attention has algebraic invariants including zero-sum rows and head-dimension rank limits, plus consistent variance spread in language models attributed to key incoherence.
-
The limits of bio-molecular modeling with large language models : a cross-scale evaluation
LLMs perform adequately on bio-molecular classification tasks but remain weak on regression, with hybrid architectures outperforming others on long sequences and fine-tuning hurting generalization.
-
Screening Is Enough
Multiscreen replaces softmax attention with screening to provide absolute query-key relevance, resulting in models with 30% fewer parameters that maintain stable performance at long contexts.
-
Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Duality
Transformers and SSMs are unified through structured state space duality, producing a 2-8X faster Mamba-2 model that remains competitive with Transformers.
-
KTO: Model Alignment as Prospect Theoretic Optimization
KTO aligns LLMs by directly maximizing prospect-theoretic utility on binary signals and matches or exceeds preference-based methods like DPO from 1B to 30B parameters.
-
Efficient Memory Management for Large Language Model Serving with PagedAttention
PagedAttention achieves near-zero waste in LLM key-value cache memory and enables 2-4x higher serving throughput than prior systems.
-
Direct Preference Optimization: Your Language Model is Secretly a Reward Model
DPO derives the optimal policy directly from human preferences via a reparameterized reward model, solving the RLHF objective with only a binary classification loss and no sampling or separate reward model.
-
QLoRA: Efficient Finetuning of Quantized LLMs
QLoRA finetunes 4-bit quantized LLMs via LoRA adapters to match full-precision performance while using far less memory, enabling 65B-scale training on single GPUs and producing Guanaco models near ChatGPT level.
-
When is Warmstarting Effective for Scaling Language Models?
A 2x growth factor in model warmstarting yields reliable training speedups for language models under 20 tokens/parameter budgets, with an empirical upper bound on effective growth factors.
-
Orthrus: Memory-Efficient Parallel Token Generation via Dual-View Diffusion
Orthrus unifies autoregressive and diffusion views on a shared KV cache to deliver lossless parallel token generation with up to 7.8x speedup and O(1) memory overhead.
-
DECO: Sparse Mixture-of-Experts with Dense-Comparable Performance on End-Side Devices
DECO matches dense model performance at 20% expert activation via ReLU-based routing with learnable scaling and the NormSiLU activation, plus a 3x real-hardware speedup.
-
Nectar: Neural Estimation of Cached-Token Attention via Regression
Nectar fits small per-layer per-head neural networks via regression to predict attention outputs and normalizers, enabling constant-time inference independent of context length while preserving semantic generation quality.
-
Dimension-Free Saddle-Point Escape in Muon
Muon achieves dimension-free saddle-point escape through non-linear spectral shaping, resolvent calculus, and structural incoherence, yielding an algebraically dimension-free escape bound.
-
LBI: Parallel Scan Backpropagation via Latent Bounded Interfaces
LBI enables tractable parallel backpropagation by reducing inter-region adjoint computation to low-dimensional r x r Jacobians while preserving exact gradients under a bounded-interface model.
-
Predicting Large Model Test Losses with a Noisy Quadratic System
A noisy quadratic system predicts large model test losses from N, B, K and outperforms Chinchilla's model for extrapolation up to 1000x compute.
-
A Communication-Theoretic Framework for LLM Agents: Cost-Aware Adaptive Reliability
LLM reliability techniques are unified as communication channel operators, with a new cost-aware router achieving superior quality-cost tradeoffs on hard tasks.
-
PRISM: Fast Online LLM Serving via Scheduling-Memory Co-design
PRISM reduces P99 TTFT by 23.3-37.1% and raises exact-prefix KV-cache hit rates by 5.9-12.2 points versus the strongest baseline on 4B and 13B models by jointly optimizing scheduling and memory.
-
Different Prompts, Different Ranks: Prompt-aware Dynamic Rank Selection for SVD-based LLM Compression
PARSE trains a prompt-aware linear router on dense-model outputs to select dynamic SVD ranks, improving accuracy up to 10% at 0.6 compression ratio on LLaMA-7B while delivering 2.5x prefill and 2.4x decode speedups.
-
Beyond Pairs: Your Language Model is Secretly Optimizing a Preference Graph
GraphDPO generalizes pairwise DPO to a graph-structured Plackett-Luce objective over DAGs induced by rollout rankings, enforcing transitivity with linear complexity and recovering DPO as a special case.
-
Bayesian Fine-tuning in Projected Subspaces
Bayesian fine-tuning of large models can be done efficiently by projecting uncertainties into low-dimensional subspaces, yielding improved calibration and generalization while keeping computational costs low.
-
Response Time Enhances Alignment with Heterogeneous Preferences
Response times modeled as drift-diffusion processes enable consistent estimation of population-average preferences from heterogeneous anonymous binary choices.
-
Can LLMs Predict Polymer Physics Just by Reading Synthesis and Processing Prose?
PolyLM fine-tunes a 9B-parameter LLM on 185k papers to predict polymer properties from text alone, achieving median R² of 0.74 on 68k held-out samples.
-
Distributional Process Reward Models: Calibrated Prediction of Future Rewards via Conditional Optimal Transport
Conditional optimal transport is used to turn raw PRM outputs into monotonic quantile functions that improve calibration and downstream Best-of-N performance on MATH-500 and AIME.
-
UniPool: A Globally Shared Expert Pool for Mixture-of-Experts
A shared global expert pool in MoE improves validation loss over per-layer experts and allows sublinear expert-parameter growth with depth.
-
Pro-KLShampoo: Projected KL-Shampoo with Whitening Recovered by Orthogonalization
Pro-KLShampoo projects KL-Shampoo preconditioners to a spike-and-flat parametric form on an r-dimensional subspace and recovers the full algebraic preconditioner via orthogonalization, outperforming KL-Shampoo on GPT-2 and LLaMA pre-training scales.