Depth-L transformers with W parameters have VC dimension Theta(L W log(T W)), yielding matching O(L W log((T+T')W)) upper and Omega(L W log((T+T')W/L)) lower bounds on sample complexity for chain-of-thought learning.
mega hub Canonical reference
LLaMA: Open and Efficient Foundation Language Models
Canonical reference. 82% of citing Pith papers cite this work as background.
1206
Pith papers citing it
Background
82% 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 206
method 19
baseline 8
other 6
dataset 1
extension 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.
mega hub controls
export citing contexts JSON
export graph JSON
export full bundle JSON
open full Pith review
annotated reader queued
Recognition alignment
counterfactual ablation
co-cited works
representative citing papers
SVHalluc benchmark shows open-source audio-visual LLMs achieve near-random accuracy on semantic and temporal speech-vision alignment tasks while Gemini 2.5 Pro performs substantially better.
Zero-Run auditing supplies valid lower bounds on differential privacy parameters from fixed member and non-member datasets by modeling and correcting distribution-shift confounding via causal-inference techniques.
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.
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.
First study of 1,899 MCP servers finds eight distinct vulnerabilities (only three traditional), 7.2% with general issues, 5.5% with tool poisoning, and 66% with code smells, urging MCP-specific security practices.
BEAVER is the first text-to-SQL benchmark from private enterprise data warehouses, revealing SOTA agentic frameworks achieve only 10.8% accuracy on complex real-world queries.
MME-RealWorld is the largest manually annotated high-resolution benchmark for MLLMs, where even the best models achieve less than 60% accuracy on challenging real-world tasks.
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.
AgentClinic is a multimodal agent benchmark demonstrating that LLM diagnostic accuracy on MedQA drops to below one-tenth in sequential clinical simulations, with Claude-3.5 leading and large tool-use differences across models.
ORPO performs preference alignment during supervised fine-tuning via a monolithic odds ratio penalty, allowing 7B models to outperform larger state-of-the-art models on alignment benchmarks.
BLaIR is a new benchmark and 570M-review dataset showing that LLM performance rankings on recommendation tasks have little correlation with rankings on general embedding benchmarks like MTEB.
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.
MMMU provides 11.5K heterogeneous college-level multimodal questions that current models solve at 56-59% accuracy, establishing a new standard for expert multimodal evaluation.
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.
API-Bank is a new benchmark and training dataset for tool-augmented LLMs that shows fine-tuned models can approach GPT-3.5 tool-use effectiveness.
GPT-4-generated instruction data produces superior zero-shot performance in finetuned LLaMA models versus prior state-of-the-art data.
Multi-agent LLMs generate and verify 14,073 deterministic reaction rules from 665,901 patents, enabling 97.7% classification of unseen reactions with finer resolution than fixed proprietary systems.
A new probing framework detects moderate parametric memorization signals in tabular in-context learning models under single-task fine-tuning, strongest on low-cardinality tasks, but signals largely disappear under realistic training.
A new sensitivity-labeled test collection is released from Enron emails with crowdsourced queries, relevance judgments, and LLM extensions for evaluating sensitivity-aware search.
LaViD distills LLM conceptual knowledge to vision models via LLM-generated MCQ soft labels, outperforming vision-language distillation baselines on fine-grained benchmarks while improving robustness on spurious correlation datasets.
citing papers explorer
-
BEAVER: An Enterprise Benchmark for Text-to-SQL
BEAVER is the first text-to-SQL benchmark from private enterprise data warehouses, revealing SOTA agentic frameworks achieve only 10.8% accuracy on complex real-world queries.
-
ORPO: Monolithic Preference Optimization without Reference Model
ORPO performs preference alignment during supervised fine-tuning via a monolithic odds ratio penalty, allowing 7B models to outperform larger state-of-the-art models on alignment benchmarks.
-
MMMU: A Massive Multi-discipline Multimodal Understanding and Reasoning Benchmark for Expert AGI
MMMU provides 11.5K heterogeneous college-level multimodal questions that current models solve at 56-59% accuracy, establishing a new standard for expert multimodal evaluation.
-
Tree of Thoughts: Deliberate Problem Solving with Large Language Models
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.
-
API-Bank: A Comprehensive Benchmark for Tool-Augmented LLMs
API-Bank is a new benchmark and training dataset for tool-augmented LLMs that shows fine-tuned models can approach GPT-3.5 tool-use effectiveness.
-
Instruction Tuning with GPT-4
GPT-4-generated instruction data produces superior zero-shot performance in finetuned LLaMA models versus prior state-of-the-art data.
-
Causally Evaluating the Learnability of Formal Language Tasks
Introduces the binning semiring and causal graphical models to show that correlational evaluation of learnability in formal language tasks leads to incorrect conclusions from confounders.
-
TrustMargin: Training-Free Arbitration between Parametric Memory and Retrieved Evidence in Large Language Models
TrustMargin arbitrates between direct and RAG answers from a frozen LLM by combining a parametric-prior margin and an evidence-binding margin computed from model likelihoods, improving results on 2WikiMQA and CWQA.
-
SurgiQ: A Large-Scale Multi-Domain Benchmark for Evaluating Surgical Understanding in Large Language Models
SurgiQ is a new 13k-question surgical benchmark showing general-purpose LLMs reach 68.1% accuracy while most biomedical models lag and smaller models stay near random baseline.
-
Parameter-Efficient Fine-Tuning with Learnable Rank
LR-LoRA learns per-layer adapter ranks during training and reports outperforming fixed-rank LoRA and other PEFT baselines on language understanding and commonsense reasoning tasks.
-
When Knowledge Is Not Free: Cost-Aware Evidence Selection in Retrieval-Augmented Generation
Defines cost-aware RAG with evidence cost tiers and shows static selectors are brittle while agentic LLM-based selection is promising but model-dependent.
-
Every Act Has Its Price: Compressed Moral Composition in Frontier LLMs
Moral Trolley Arena shows frontier LLMs produce composite moral preferences that are compressed rather than additive functions of calibrated component act strengths across Moral Foundations Theory.
-
Phantom transitions in language model fine-tuning
Apparent phase transitions during fine-tuning on near-synonym tasks are phantoms originating in the softmax readout; an order parameter isolates kinematic and structural failure modes and a few dimensionless quantities predict critical learning rates across architectures via blind test.
-
Large Language Model Selection with Limited Annotations
SELECT-LLM is the first active model selection framework for LLMs that uses expected information gain from pairwise output similarities to minimize required annotations, reporting up to 84.78% cost reduction across 23 datasets and 156 models.
-
Brain-LLM Alignment Tracks Training Data, Not Typology
Training-language dominance, not English inherent properties, determines brain-LLM alignment across English, Chinese, and French, with additional independent effects from typological distance concentrated in syntactic brain regions.
-
Understanding Data Temporality Impact on Large Language Models Pre-training
Pre-training 6B LLMs on temporally ordered Common Crawl snapshots yields models with improved factual freshness and temporal precision over shuffled baselines while matching on general language understanding.
-
Dynamic Chunking for Diffusion Language Models
DCDM replaces positional blocks with learnable semantic chunks via differentiable Chunking Attention, yielding consistent gains over block and unstructured diffusion baselines up to 1.5B parameters.
-
IndicMedDialog: A Parallel Multi-Turn Medical Dialogue Dataset for Accessible Healthcare in Indic Languages
A parallel multi-turn medical dialogue dataset spanning English and nine Indic languages is created from synthetic consultations to enable personalized AI healthcare interactions.
-
GKnow: Measuring the Entanglement of Gender Bias and Factual Gender
Gender bias and factual gender knowledge are severely entangled in language model circuits and neurons, making neuron ablation an unreliable method for debiasing.
-
Learning Less Is More: Premature Upper-Layer Attention Specialization Hurts Language Model Pretraining
Temporarily reducing the learning rate on upper-layer query and key projections during early GPT pretraining prevents premature attention specialization and improves model performance.
-
Chain-based Distillation for Effective Initialization of Variable-Sized Small Language Models
Chain-based Distillation constructs a sequence of anchor models to enable efficient initialization of variable-sized SLMs through interpolation, with bridge distillation for cross-architecture transfer, yielding better performance than scratch training.
-
Understanding Performance Collapse in Layer-Pruned Large Language Models via Decision Representation Transitions
Performance collapse in layer-pruned LLMs stems from disrupting the Silent Phase of decision-making, which blocks the transition to correct predictions, while the later Decisive Phase is robust to pruning.
-
MIST: Multimodal Interactive Speech-based Tool-calling Conversational Assistants for Smart Homes
MIST is a new synthetic speech-based tool-calling dataset for IoT devices that exposes performance gaps between open- and closed-weight multimodal LLMs.
-
VITA-QinYu: Expressive Spoken Language Model for Role-Playing and Singing
VITA-QinYu is the first expressive end-to-end spoken language model supporting role-playing and singing alongside conversation, trained on 15.8K hours of data and outperforming prior models on expressiveness and conversational benchmarks.
-
NoisyCausal: A Benchmark for Evaluating Causal Reasoning Under Structured Noise
NoisyCausal benchmark tests LLMs on causal reasoning with structured noise, and a modular LLM-plus-causal-graph framework outperforms baselines while generalizing to Cladder.
-
VIDA: A dataset for Visually Dependent Ambiguity in Multimodal Machine Translation
VIDA provides 2,500 visually-dependent ambiguous translation examples and span-level disambiguation metrics; CoT-SFT on LVLMs improves out-of-distribution performance over standard SFT.
-
A Multi-View Media Profiling Suite: Resources, Evaluation, and Analysis
Presents MBFC-2025 dataset and multi-view embeddings with fusion methods for media bias and factuality, reporting SOTA results on ACL-2020 and new benchmarks on MBFC-2025.
-
Quantifying Prior Dominance in RAG Systems
Introduces NCU metric using token log-probabilities and finds small language models match or outperform larger ones in strict factual RAG extraction, while commercial APIs show high prior dominance and negative transfer.
-
Fine-tuning vs. In-context Learning in Large Language Models: A Formal Language Learning Perspective
A controlled formal language task reveals fine-tuning outperforms in-context learning on in-distribution generalization but equals it on out-of-distribution, with ICL showing greater sensitivity to model size and tokenization.
-
Misinformation Span Detection in Videos via Audio Transcripts
New datasets and language model classifiers enable detection of misinformation spans in video transcripts with an F1 score of 0.68.
-
Cognitive Policy-Driven LLM for Diagnosis and Intervention of Cognitive Distortions in Emotional Support Conversation
The CogBiasESC dataset and CoPoLLM framework enable LLMs to diagnose cognitive distortions and apply interventions in emotional support conversations, outperforming baselines on accuracy, effectiveness, and safety.
-
Modeling Multi-Dimensional Cognitive States in Large Language Models under Cognitive Crowding
CognitiveBench reveals LLMs suffer representation overlap on joint cognitive tasks due to hierarchical structure; HyCoLLM in hyperbolic space fixes the mismatch and outperforms GPT-4o with far fewer parameters.
-
RAGognizer: Hallucination-Aware Fine-Tuning via Detection Head Integration
RAGognizer adds a detection head to LLMs for joint training on generation and token-level hallucination detection, yielding SOTA detection and fewer hallucinations in RAG while preserving output quality.
-
Robust Explanations for User Trust in Enterprise NLP Systems
Decoder LLMs produce substantially more stable explanations than encoder models, with 73% lower top-token flip rates on average and further 44% gains from 7B to 70B scale.
-
MIXAR: Scaling Autoregressive Pixel-based Language Models to Multiple Languages and Scripts
MIXAR is the first autoregressive pixel-based language model for eight languages and scripts, with empirical gains on multilingual tasks, robustness to unseen languages, and further improvements when scaled to 0.5B parameters.
-
In-Context Learning in Speech Language Models: Analyzing the Role of Acoustic Features, Linguistic Structure, and Induction Heads
Speech language models show in-context learning where speaking rate affects both accuracy and mimicry, and induction heads are causally necessary for this capability.
-
OmniCompliance-100K: A Multi-Domain, Rule-Grounded, Real-World Safety Compliance Dataset
OmniCompliance-100K supplies 12,985 distinct rules and 106,009 associated real-world cases from 74 multi-domain regulations to benchmark LLM safety and compliance.
-
Disco-RAG: Discourse-Aware Retrieval-Augmented Generation
Disco-RAG improves RAG by building intra-chunk discourse trees and inter-chunk rhetorical graphs that feed into a planning blueprint, delivering state-of-the-art results on question answering and long-document summarization without any fine-tuning.
-
Hearing to Translate: The Effectiveness of Speech Modality Integration into LLMs
Cascaded systems remain the most reliable for speech translation overall, but recent SpeechLLMs match or outperform them in many conditions while standalone speech models lag.
-
LogitTrace: Detecting Benchmark Contamination via Layerwise Logit Trajectories
LogitTrace detects benchmark contamination by showing that contaminated inputs produce earlier stabilization in layerwise logit trajectories while clean inputs show more gradual accumulation.
-
PuzzleWorld: A Benchmark for Multimodal, Open-Ended Reasoning in Puzzlehunts
PuzzleWorld benchmark reveals state-of-the-art AI models solve only 18% of complex puzzlehunt problems with 40% stepwise accuracy, matching novices but trailing enthusiasts, while fine-tuning on traces yields modest gains.
-
Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression
KV cache compression causes task-dependent degradation in high-density reasoning due to disrupted CoT links; ShotKV mitigates this by preserving few-shot examples as indivisible semantic units through phase separation, delivering 9-18% accuracy gains and 11% latency reduction.
-
Do NOT Think That Much for 2+3=? On the Overthinking of o1-Like LLMs
o1-like models overthink easy tasks; self-training reduces compute use without accuracy loss on GSM8K, MATH500, GPQA, and AIME.
-
Federated Co-tuning Framework for Large and Small Language Models
FedCoLLM is a parameter-efficient federated co-tuning framework that improves client SLMs via server LLMs and enriches LLMs with client domain insights using adapters on NLP text generation tasks.
-
DuoAttention: Efficient Long-Context LLM Inference with Retrieval and Streaming Heads
DuoAttention identifies retrieval heads requiring full KV cache and streaming heads using constant-length cache to reduce memory and latency in long-context LLM inference.
-
DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model
DeepSeek-V2 delivers top-tier open-source LLM performance using only 21B active parameters by compressing the KV cache 93.3% and cutting training costs 42.5% via MLA and DeepSeekMoE.
-
The Era of 1-bit LLMs: All Large Language Models are in 1.58 Bits
BitNet b1.58 shows that ternary 1.58-bit LLMs can match full-precision performance at substantially lower inference cost.
-
Hallucination is Inevitable: An Innate Limitation of Large Language Models
Hallucinations are inevitable in LLMs because they cannot learn all computable functions according to learning theory.
-
Self-Rewarding Language Models
Iterative self-rewarding via LLM-as-Judge in DPO training on Llama 2 70B improves instruction following and self-evaluation, outperforming GPT-4 on AlpacaEval 2.0.
-
Detecting Pretraining Data from Large Language Models
Min-K% Prob detects pretraining data in LLMs by flagging outlier low-probability words in text, achieving 7.4% better performance than prior methods on the new WIKIMIA benchmark.