Presents a new expert-curated dataset of multi-turn counterspeech dialogues in five languages targeting hate against seven groups, with span annotations linking to verified external knowledge for RAG applications.
super hub Canonical reference
Efficient Estimation of Word Representations in Vector Space
Canonical reference. 75% of citing Pith papers cite this work as background.
187
Pith papers citing it
18.1k
external citations · Pith
Background
75% of classified citations
abstract
We propose two novel model architectures for computing continuous vector representations of words from very large data sets. The quality of these representations is measured in a word similarity task, and the results are compared to the previously best performing techniques based on different types of neural networks. We observe large improvements in accuracy at much lower computational cost, i.e. it takes less than a day to learn high quality word vectors from a 1.6 billion words data set. Furthermore, we show that these vectors provide state-of-the-art performance on our test set for measuring syntactic and semantic word similarities.
hub tools
citation-role summary
background 22
method 5
dataset 1
citation-polarity summary
claims ledger
- abstract We propose two novel model architectures for computing continuous vector representations of words from very large data sets. The quality of these representations is measured in a word similarity task, and the results are compared to the previously best performing techniques based on different types of neural networks. We observe large improvements in accuracy at much lower computational cost, i.e. it takes less than a day to learn high quality word vectors from a 1.6 billion words data set. Furthermore, we show that these vectors provide state-of-the-art performance on our test set for measuri
authors
co-cited works
representative citing papers
GPT-3 shows that scaling an autoregressive language model to 175 billion parameters enables strong few-shot performance across diverse NLP tasks via in-context prompting without fine-tuning.
REALM augments language-model pre-training with an unsupervised retriever over Wikipedia documents and reports 4-16% absolute gains on open-domain QA benchmarks over prior implicit and explicit knowledge methods.
Deep neural networks exhibit distributed high-level semantic representations and discontinuous input-output mappings vulnerable to transferable adversarial perturbations.
OntoLearner supplies the first cross-domain ontology collection and benchmarking infrastructure for LLM-driven ontology learning, finding that failure scales with ontological complexity instead of model size.
GRPO, Dr. GRPO, and DAPO are three settings of one dial on the group standard deviation of binary rewards, unified by the group-standard-deviation identity where disagreement equals update magnitude.
The paper defines STE and SPID, two information-theoretic measures of semantic flow and decomposition in language exchanges, and applies them to four dialogue datasets.
A Set-Transformer architecture with self-attention encodes Pauli-string correlations, optimizes via commutation objective, and finds symmetries with near-deterministic success on physical models like Ising and Toric code.
Proposes latent analogies and analogy transduction to enable compositional generalization to unseen goal-context pairs in offline GCRL, outperforming trajectory-stitching baselines on manipulation tasks.
PluRule is a new multimodal multilingual benchmark showing that state-of-the-art vision-language models perform only marginally better than a trivial baseline at detecting specific rule violations in pluralistic online communities.
Proposes Wasserstein Projection Mechanism for differentially private sampling that optimizes Wasserstein distance utility and provides convergence guarantees for approximate computation.
Defines OZ-TAL task and presents a training-free VLM-based method that outperforms prior approaches for online and offline zero-shot temporal action localization on THUMOS14 and ActivityNet-1.3.
Hybrid human-AI networks in 5x5 grids reached lower final polarization than human-only networks after eight rounds of opinion revision on polarizing topics.
EditRefiner uses a perception-reasoning-action-evaluation agent loop and the EditFHF-15K human feedback dataset to refine text-guided image edits more accurately than prior methods.
Autoregressive semantic ID generation creates tree-induced probability correlations that prevent generative recommenders from capturing simple patterns; Latte adds latent tokens to relax these correlations.
Using historical corpora and the Rational Speech Act framework, attested English morphological compositions are ranked higher than plausible alternatives from the same time period when both semantic recoverability and production cost are considered.
EPIC trains LLMs to treat continuous embeddings as in-context prompts, yielding state-of-the-art text embedding performance on MTEB with or without prompts at inference and lower compute.
A code-and-comment analysis method detects semantic clones in Solidity functions with 59% overall precision (84% for same-name functions) and 97% recall on 300k contracts, plus LLM summaries for uncommented code.
PROVFUSION fuses three complementary views of provenance data with lightweight schemes and voting to achieve higher detection accuracy and lower false positives than node- or edge-only baselines on nine benchmarks.
TTT-Discover applies test-time RL to set new state-of-the-art results on math inequalities, GPU kernels, algorithm contests, and single-cell denoising using an open model and public code.
GRAB is a benchmark dataset of 1.61M sentences from 8,247 10-K filings with taxonomy-anchored weak supervision labels for standardized evaluation of unsupervised topic models on financial risk disclosures.
A survey of LLM copyright protection that unifies text watermarking, model watermarking, and model fingerprinting while presenting new coverage of fingerprint transfer and removal.
Pioneers ViPro, the first attack to adversarially promote videos in text-to-video retrieval, using Modal Refinement to improve black-box transferability across multiple targets.
VLM2Vec converts state-of-the-art vision-language models into universal multimodal embedders via contrastive training on the new MMEB benchmark, delivering 10-20% absolute gains over prior models on both in-distribution and out-of-distribution tasks.
citing papers explorer
-
CATCH-ME if you RAG: a dataset of Contextually Annotated multi-Turn Counterspeech against Hate and Misinformation Exchanges
Presents a new expert-curated dataset of multi-turn counterspeech dialogues in five languages targeting hate against seven groups, with span annotations linking to verified external knowledge for RAG applications.
-
Language Models are Few-Shot Learners
GPT-3 shows that scaling an autoregressive language model to 175 billion parameters enables strong few-shot performance across diverse NLP tasks via in-context prompting without fine-tuning.
-
REALM: Retrieval-Augmented Language Model Pre-Training
REALM augments language-model pre-training with an unsupervised retriever over Wikipedia documents and reports 4-16% absolute gains on open-domain QA benchmarks over prior implicit and explicit knowledge methods.
-
Intriguing properties of neural networks
Deep neural networks exhibit distributed high-level semantic representations and discontinuous input-output mappings vulnerable to transferable adversarial perturbations.
-
OntoLearner: A Modular Python Library for Ontology Learning with Large Language Models
OntoLearner supplies the first cross-domain ontology collection and benchmarking infrastructure for LLM-driven ontology learning, finding that failure scales with ontological complexity instead of model size.
-
GRPO, Dr. GRPO, and DAPO Are Three Operations on One Number: The Group-Standard-Deviation Identity
GRPO, Dr. GRPO, and DAPO are three settings of one dial on the group standard deviation of binary rewards, unified by the group-standard-deviation identity where disagreement equals update magnitude.
-
Information Dynamics of Language Communication
The paper defines STE and SPID, two information-theoretic measures of semantic flow and decomposition in language exchanges, and applies them to four dialogue datasets.
-
Attention-based optimizer for symmetry finding
A Set-Transformer architecture with self-attention encodes Pauli-string correlations, optimizes via commutation objective, and finds symmetries with near-deterministic success on physical models like Ising and Toric code.
-
Compositional Transduction with Latent Analogies for Offline Goal-Conditioned Reinforcement Learning
Proposes latent analogies and analogy transduction to enable compositional generalization to unseen goal-context pairs in offline GCRL, outperforming trajectory-stitching baselines on manipulation tasks.
-
PluRule: A Benchmark for Moderating Pluralistic Communities on Social Media
PluRule is a new multimodal multilingual benchmark showing that state-of-the-art vision-language models perform only marginally better than a trivial baseline at detecting specific rule violations in pluralistic online communities.
-
Differentially Private Sampling from Distributions via Wasserstein Projection
Proposes Wasserstein Projection Mechanism for differentially private sampling that optimizes Wasserstein distance utility and provides convergence guarantees for approximate computation.
-
OZ-TAL: Online Zero-Shot Temporal Action Localization
Defines OZ-TAL task and presents a training-free VLM-based method that outperforms prior approaches for online and offline zero-shot temporal action localization on THUMOS14 and ActivityNet-1.3.
-
An Experimental Method to Study Opinion Diffusion in Human-AI Hybrid Societies
Hybrid human-AI networks in 5x5 grids reached lower final polarization than human-only networks after eight rounds of opinion revision on polarizing topics.
-
EditRefiner: A Human-Aligned Agentic Framework for Image Editing Refinement
EditRefiner uses a perception-reasoning-action-evaluation agent loop and the EditFHF-15K human feedback dataset to refine text-guided image edits more accurately than prior methods.
-
Expressiveness Limits of Autoregressive Semantic ID Generation in Generative Recommendation
Autoregressive semantic ID generation creates tree-induced probability correlations that prevent generative recommenders from capturing simple patterns; Latte adds latent tokens to relax these correlations.
-
Rational Communication Shapes Morphological Composition
Using historical corpora and the Rational Speech Act framework, attested English morphological compositions are ranked higher than plausible alternatives from the same time period when both semantic recoverability and production cost are considered.
-
Embedding-based In-Context Prompt Training for Enhancing LLMs as Text Encoders
EPIC trains LLMs to treat continuous embeddings as in-context prompts, yielding state-of-the-art text embedding performance on MTEB with or without prompts at inference and lower compute.
-
Identifying and Characterizing Semantic Clones of Solidity Functions
A code-and-comment analysis method detects semantic clones in Solidity functions with 59% overall precision (84% for same-name functions) and 97% recall on 300k contracts, plus LLM summaries for uncommented code.
-
Beyond Nodes vs. Edges: A Multi-View Fusion Framework for Provenance-Based Intrusion Detection
PROVFUSION fuses three complementary views of provenance data with lightweight schemes and voting to achieve higher detection accuracy and lower false positives than node- or edge-only baselines on nine benchmarks.
-
Learning to Discover at Test Time
TTT-Discover applies test-time RL to set new state-of-the-art results on math inequalities, GPU kernels, algorithm contests, and single-cell denoising using an open model and public code.
-
GRAB: A Risk Taxonomy--Grounded Benchmark for Unsupervised Topic Discovery in Financial Disclosures
GRAB is a benchmark dataset of 1.61M sentences from 8,247 10-K filings with taxonomy-anchored weak supervision labels for standardized evaluation of unsupervised topic models on financial risk disclosures.
-
Copyright Protection for Large Language Models: A Survey of Methods, Challenges, and Trends
A survey of LLM copyright protection that unifies text watermarking, model watermarking, and model fingerprinting while presenting new coverage of fingerprint transfer and removal.
-
Adversarial Video Promotion Against Text-to-Video Retrieval
Pioneers ViPro, the first attack to adversarially promote videos in text-to-video retrieval, using Modal Refinement to improve black-box transferability across multiple targets.
-
VLM2Vec: Training Vision-Language Models for Massive Multimodal Embedding Tasks
VLM2Vec converts state-of-the-art vision-language models into universal multimodal embedders via contrastive training on the new MMEB benchmark, delivering 10-20% absolute gains over prior models on both in-distribution and out-of-distribution tasks.
-
A Simple Framework for Contrastive Learning of Visual Representations
SimCLR learns visual representations by contrasting augmented views of the same image and reaches 76.5% ImageNet top-1 accuracy with a linear classifier, matching a supervised ResNet-50.
-
BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension
BART introduces a denoising pretraining method for seq2seq models that matches RoBERTa on GLUE and SQuAD while setting new state-of-the-art results on abstractive summarization, dialogue, and QA with up to 6 ROUGE gains.
-
Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer
T5 casts all NLP tasks as text-to-text generation, systematically explores pre-training choices, and reaches strong performance on summarization, QA, classification and other tasks via large-scale training on the Colossal Clean Crawled Corpus.
-
Language Models as Knowledge Bases?
BERT stores relational knowledge extractable via cloze queries without fine-tuning and matches supervised baselines on open-domain QA tasks.
-
Hindi Visual Genome: A Dataset for Multimodal English-to-Hindi Machine Translation
The paper releases the first multimodal English-Hindi machine translation dataset of 31,525 segments with images and a challenge test set of 1,400 segments selected via embedding similarity for image-resolvable ambiguities.
-
Tight Sensitivity Bounds For Smaller Coresets
New algorithms compute provably tight sensitivity bounds for matrix rows, yielding smaller coresets for LMS approximation of affine k-subspaces via an iterative exact method and a dimensionality-reduction trick.
-
From Execution to Education: A Bloom-Aligned Framework for Measuring Educational Control in LLMs
On 2,520 programming tasks, matched Qwen general and coder models reliably raise Bloom cognitive demand but fail to lower it, so execution skill does not imply educational control.
-
Prompt Compression via Activation Aggregation
A learned weighted sum of intermediate-layer activations compresses an instruction prompt into a single patch vector that, injected at an early layer, recovers task accuracy within ~2% of the full prompt.
-
Recovering Latent Structures after Variational Bayesian Variable Selection: Fit Assessment and Factor-Number Selection in Partially Exploratory Factor Analysis
A scale-free gain rule applied to variational ELBO paths recovers true factor dimensionality in partially exploratory factor analysis where raw information criteria over-factor.
-
From Text to Parameters: Predicting Item Parameters from Embedding Regularization with Reliability and Design Ceilings
Text embeddings recover 57-63% of the reliable variance in exam-item difficulty, and apparent differences in predictability across IRT parameters are mostly artifacts of calibration noise rather than text signal.
-
FedLAB: Traceable Semantic Codebooks for Federated Multimodal Graph Foundation Learning
FedLAB organizes multimodal graph knowledge into typed hierarchical codebooks for modality evidence, node semantics, and topology context via federated semantic barycenter pre-training, improving performance by up to 7.53% on benchmarks while enabling semantic traceability.
-
Residual-Guided Expert Specialization for Incomplete Multimodal Learning
MARS is a residual-guided MoE framework for incomplete multimodal learning that uses training-time residuals to train a deployable feature router and adds discrepancy-aware noise regularization.
-
Radical AI Interpretability
A framework is proposed for solving for an AI system's beliefs and desires from its computational facts, with criteria for success tied to interpretability tests and emphasis on holistic attribution.
-
Behavioral and Representational Evidence of Binomial Ordering Preferences in Large Language Models
LLMs recover dominant binomial orders from corpora but align less closely with exact preference distributions, with preference strength partially encoded in middle-to-late layers and manipulable via steering.
-
Disentanglement with Holographic Reduced Representations
The paper proposes an HRR-based unsupervised method for disentanglement, proves that the unbinding operation induces approximately independent symbol-value pairs, and derives a per-slot capacity bound.
-
When More Cores Hurts: The Vector Database Scaling Paradox in HPC
Large-scale HPC evaluation of Qdrant, Milvus, and Weaviate reveals that workload patterns limit scaling and extra cores can reduce throughput, exposing a cloud-to-HPC design mismatch.
-
Larch: Learned Query Optimization for Semantic Predicates
Larch uses a GNN-MDP formulation and a selectivity predictor plus dynamic programming to reorder semantic filter evaluation, cutting token usage 3x-19x versus prior systems on real and synthetic workloads.
-
LLM Self-Recognition: Steering and Retrieving Activation Signatures
Steering LLM residual streams with random sparse vectors creates detectable self-recognition fingerprints that enable over 98% accurate attribution of generated text to specific models without degrading output quality.
-
Graph Cascades: Contagion-Based Mesoscopic Rewiring for Structure-Aware Graph Machine Learning
Graph Cascades uses contagion diffusion to rewire graphs by promoting reinforced multi-hop node pairs to direct neighbors, improving GNN performance on heterophilic and moderate-degree homophilic graphs under specified conditions.
-
Intra-Modal Neighbors Never Lie: Rectifying Inter-Modal Noisy Correspondence via Graph-Based Intra-Modal Reasoning
IN2R rectifies inter-modal noisy correspondence by synthesizing continuous soft prototypes from intra-modal neighbor consensus using a Graph Refiner on dynamic cross-modal memory.
-
Pay Beliefs and the Amenity-Pay Tradeoff
Biased pay beliefs distort amenity-pay tradeoffs in stated preferences, with full disclosure restoring alignment to full-info benchmarks while short-term info does not.
-
Evolving and Detecting Multi-Turn Deception using Geometric Signatures
Multi-objective genetic prompt optimization creates multi-turn deceptive datasets validated by humans, then detected with 0.89 recall using angular coverage, distance ratio, and linearity features in embeddings.
-
HCL-FF: Hierarchical and Contrastive Learning for Forward-Forward Algorithm
HCL-FF augments the Forward-Forward algorithm with hierarchical learning and contrastive objectives to reach new state-of-the-art accuracies among FF methods on CIFAR-10 (+5.46%), CIFAR-100 (+17.00%), and Tiny-ImageNet (+12.51%).
-
Deep-Research Agents Can Be Poisoned via User-Generated Content
Deep-research agents have a concentrated attack surface because they repeatedly retrieve the same UGC pages, allowing a single poisoned page to affect citations and entity promotion across query clusters in systems like STORM, Co-STORM, and OmniThink.
-
Dynamic Query Modification for Binary Locality Sensitive Hashing
Dynamic query modification approximates the ℓ2-normalised centroid of k-nearest neighbours to raise binary LSH collision probability and reduce ANN search cost in MQ-Forest.
-
Assisted Counterspeech Writing at the Crossroads of Hate Speech and Misinformation
LLMs generate adequate counterspeech for co-occurring hate and misinformation in 40% of cases, with a mixed knowledge strategy from fact-checkers and NGOs proving most effective after expert revision.