Introduces unified span-level hallucination detection benchmark over code, tool output, and documents; fine-tuned Qwen3.5-2B reaches 0.689 span-F1 and outperforms baselines including on code-agent data.
hub
LFM2 technical report.arXiv preprint arXiv:2511.23404
19 Pith papers cite this work. Polarity classification is still indexing.
19
Pith papers citing it
hub tools
citation-role summary
background 2
baseline 1
contradiction 1
citation-polarity summary
representative citing papers
NetLLMeval is an emulation-based framework for benchmarking LLM solvers on network admin tasks, with a 24000-run study showing solver architecture lifts a 14B model from 0.43 to 0.88 accuracy and allows local models to match frontier systems.
LlamaWeb is a WebGPU backend for llama.cpp that uses static memory planning, tunable kernels, and templated multi-precision support to cut memory use by 29-33% and raise decode throughput by 45-69% versus prior browser frameworks on tested hardware.
Preconditioned delta-rule models with a diagonal curvature approximation improve upon standard DeltaNet, GDN, and KDA by better approximating the test-time regression objective.
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.
Alignment of vision-language models with human V1-V3 early visual cortex negatively predicts resistance to sycophantic gaslighting attacks.
Agreeableness in AI personas reliably predicts sycophantic behavior in 9 of 13 tested language models.
TiCo enables spoken dialogue models to follow explicit time constraints in generated responses using Spoken Time Markers and reinforcement learning with verifiable rewards, cutting duration error by 2.7x over its backbone.
PluraMath extends PolyMath with human-validated math problems in 18 mid-to-extreme low-resource languages and benchmarks 27 reasoning LLMs, finding a persistent high- vs low-resource performance gap.
MLAIRE is a protocol that evaluates multilingual retrievers on both semantic accuracy and query-language preference using parallel passages and new metrics like LPR and Lang-nDCG, showing that standard metrics hide distinct behavioral differences among retrievers.
6G networks need LLM-based agents in a layered semantic control plane to achieve autonomous intelligence, with empirical results showing that heterogeneous deployment across device-edge-core is required due to inherent tradeoffs in reasoning, latency, and efficiency.
LEMUR accelerates multi-vector retrieval by learning a neural network approximation to MaxSim and reducing it to single-vector search in latent space.
SHARe-KAN compresses KAN prediction-head storage by 9.3X via post-training vector quantization at a 2-point mAP cost on PASCAL VOC detection, with no retraining and good zero-shot transfer.
OCC-RAG develops task-specialized SLMs (0.6B and 1.7B) via a new synthetic data pipeline for multi-hop reasoning and context faithfulness, claiming to match or exceed 2-6x larger general models on HotpotQA, MuSiQue, TAT-QA, ConFiQA, and MuSiQue-Un.
SafeLens presents a fast-and-slow video guardrail framework that filters the SafeWatch dataset to 2.4% and adds Chain-of-Thought traces to achieve state-of-the-art moderation performance at reduced inference cost.
MeMo encodes new knowledge into a separate memory model that integrates with frozen LLMs, showing strong performance on QA benchmarks while avoiding catastrophic forgetting and working without access to model weights.
A 1.2B-parameter open-weight swarm with hand-crafted scaffolding reproduces jailbreak and 9-CWE vulnerability-discovery capabilities attributed to restricted frontier models.
A multi-agent architecture using compact VLMs on edge hardware controls a mobile manipulator across five warehouse task categories in hardware-in-the-loop simulation.
citing papers explorer
-
Beyond Document Grounding: Span-Level Hallucination Detection over Code, Tool Output, and Documents
Introduces unified span-level hallucination detection benchmark over code, tool output, and documents; fine-tuned Qwen3.5-2B reaches 0.689 span-F1 and outperforms baselines including on code-agent data.
-
Toward Agentic SysAdmin: Rethinking System Administration with AI Agents
NetLLMeval is an emulation-based framework for benchmarking LLM solvers on network admin tasks, with a 24000-run study showing solver architecture lifts a 14B model from 0.43 to 0.88 accuracy and allows local models to match frontier systems.
-
Llamas on the Web: Memory-Efficient, Performance-Portable, and Multi-Precision LLM Inference with WebGPU
LlamaWeb is a WebGPU backend for llama.cpp that uses static memory planning, tunable kernels, and templated multi-precision support to cut memory use by 29-33% and raise decode throughput by 45-69% versus prior browser frameworks on tested hardware.
-
Preconditioned DeltaNet: Curvature-aware Sequence Modeling for Linear Recurrences
Preconditioned delta-rule models with a diagonal curvature approximation improve upon standard DeltaNet, GDN, and KDA by better approximating the test-time regression objective.
-
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.
-
Gaslight, Gatekeep, V1-V3: Early Visual Cortex Alignment Shields Vision-Language Models from Sycophantic Manipulation
Alignment of vision-language models with human V1-V3 early visual cortex negatively predicts resistance to sycophantic gaslighting attacks.
-
Too Nice to Tell the Truth: Quantifying Agreeableness-Driven Sycophancy in Role-Playing Language Models
Agreeableness in AI personas reliably predicts sycophantic behavior in 9 of 13 tested language models.
-
TiCo: Time-Controllable Spoken Dialogue Model
TiCo enables spoken dialogue models to follow explicit time constraints in generated responses using Spoken Time Markers and reinforcement learning with verifiable rewards, cutting duration error by 2.7x over its backbone.
-
PluraMath: Extending Mathematical Reasoning Evaluation Beyond High-Resource Languages
PluraMath extends PolyMath with human-validated math problems in 18 mid-to-extreme low-resource languages and benchmarks 27 reasoning LLMs, finding a persistent high- vs low-resource performance gap.
-
MLAIRE: Multilingual Language-Aware Information Retrieval Evaluation Protocal
MLAIRE is a protocol that evaluates multilingual retrievers on both semantic accuracy and query-language preference using parallel passages and new metrics like LPR and Lang-nDCG, showing that standard metrics hide distinct behavioral differences among retrievers.
-
6G Needs Agents: Toward Agentic AI-Native Networks for Autonomous Intelligence
6G networks need LLM-based agents in a layered semantic control plane to achieve autonomous intelligence, with empirical results showing that heterogeneous deployment across device-edge-core is required due to inherent tradeoffs in reasoning, latency, and efficiency.
-
LEMUR: Learned Multi-Vector Retrieval
LEMUR accelerates multi-vector retrieval by learning a neural network approximation to MaxSim and reducing it to single-vector search in latent space.
-
SHARe-KAN: Post-Training Vector Quantization for Cache-Resident KAN Inference
SHARe-KAN compresses KAN prediction-head storage by 9.3X via post-training vector quantization at a 2-point mAP cost on PASCAL VOC detection, with no retraining and good zero-shot transfer.
-
OCC-RAG: Optimal Cognitive Core for Faithful Question Answering
OCC-RAG develops task-specialized SLMs (0.6B and 1.7B) via a new synthetic data pipeline for multi-hop reasoning and context faithfulness, claiming to match or exceed 2-6x larger general models on HotpotQA, MuSiQue, TAT-QA, ConFiQA, and MuSiQue-Un.
-
SafeLens: Deliberate and Efficient Video Guardrails with Fast-and-Slow Screening
SafeLens presents a fast-and-slow video guardrail framework that filters the SafeWatch dataset to 2.4% and adds Chain-of-Thought traces to achieve state-of-the-art moderation performance at reduced inference cost.
-
MeMo: Memory as a Model
MeMo encodes new knowledge into a separate memory model that integrates with frozen LLMs, showing strong performance on QA benchmarks while avoiding catastrophic forgetting and working without access to model weights.
-
AI Security Policy Should Assess Systems, Not Only Models
A 1.2B-parameter open-weight swarm with hand-crafted scaffolding reproduces jailbreak and 9-CWE vulnerability-discovery capabilities attributed to restricted frontier models.
-
Multi-Agent Robotic Control with Onboard Vision-Language Models
A multi-agent architecture using compact VLMs on edge hardware controls a mobile manipulator across five warehouse task categories in hardware-in-the-loop simulation.
- ATMA: Long-Context Language Modeling via Polar Attention and Gated-Delta Compression Memory