Results for “llm-bias”
56 skills
qhjqhj00 Cab Eval
Benchmarks LLM bias by scoring responses to automatically generated open-ended questions across sensitive attributes, producing a composite fitness score from 0 to 5.
3
muratcankoylan Advanced Evaluation
Provides production-grade techniques for evaluating LLM outputs using LLMs as judges, covering direct scoring, pairwise comparison, bias mitigation, rubric generation, and confidence calibration.
16.9k · bundle
shulkwisec LLM Testing
Comprehensive LLM security testing prompts for bias detection, data leakage, alignment testing, and adversarial prompt resistance.
21 · bundle
dvy1987 Eval Judge
Score LLM and agent outputs using LLM-as-judge techniques — direct scoring against rubrics or pairwise comparison between two outputs. Includes built-in bias mitigation for position bias, length bias, and self-enhancement bias. Load when the user asks to score an output, judge a response, evaluate against a rubric, compare two outputs, do direct scoring, run pairwise comparison, or says "rate this", "which response is better", "score this against the rubric", "judge this output", "LLM as judge this". Sub-skill of eval-output orchestrator.
3 · bundle
kursku Advanced Evaluation
This skill should be used when the user asks to "implement LLM-as-judge", "compare model outputs", "create evaluation rubrics", "mitigate evaluation bias", or mentions direct scoring, pairwise comparison, position bias, evaluation pipelines, or automated quality assessment.
55 · bundle
More results
projectious-work LLM Evaluation
LLM output evaluation — automated metrics, LLM-as-judge, A/B testing, regression testing. Use when measuring LLM output quality, comparing prompt or model versions, building an automated eval pipeline, setting up regression tests for prompt changes, or evaluating RAG systems and bias/safety.
0
neuralblitz Llmops
Manages the lifecycle of large language models in production, covering model versioning, prompt management, inference optimization, and cost control.
1
jiachen-t-wang Lima Less Is More For Alignment Arxiv 2305 11206v1
LIMA: Less Is More for Alignment
6
snoodleboot-io Cloud Iam Design
Least privilege is easy to state and hard to hold: grant each identity only the
2
jiachen-t-wang Sigmoid Loss For Language Image Pre Training Arxiv 2303 1534
Sigmoid Loss for Language Image Pre-Training
6
drnabeelkhan L3 2 Legal Vertical
Applies ABA Model Rules of Professional Conduct to enforce attorney-client privilege, ethics walls, and conflict-of-interest patterns in legal workflows.
2
qcmuu Llamaguard
Meta's 7-8B specialized moderation model for LLM input/output filtering. 6 safety categories - violence/hate, sexual content, weapons, substances, self-harm, criminal planning. 94-95% accuracy. Deploy with vLLM, HuggingFace, Sagemaker. Integrates with NeMo Guardrails.
0
pranavnagrecha Lwc Pubsub Patterns
Implements sibling communication between Lightning Web Components using Lightning Message Service (LMS) or the legacy pubsub utility, covering message channel definition, publish/subscribe patterns, and scope selection.
15 · bundle
drnabeelkhan Maxim Banner Design
Creates digital banners for social, display, hero, and email formats, applying behavioral science principles to maximize attention and conversion.
2 · bundle
nvidia Tao Analyze Gaps Vlm Bcq
Extract false-positive and false-negative gaps from VLM binary-classification-question predictions by comparing model responses against ground truth, producing a structured JSONL file and summary report for downstream root-cause analysis.
2.2k · bundle
lambenthan Review
通用跨模型审查：Review LLM 对任意研究制品进行独立评审，输出结构化评分、wiki 实体映射与改进建议
77
qcmuu Blip 2 Vision Language
Vision-language pre-training framework bridging frozen image encoders and LLMs. Use when you need image captioning, visual question answering, image-text retrieval, or multimodal chat with state-of-the-art zero-shot performance.
0 · bundle
nexu-io Frame Light Leak Cinema
Generates a single-frame HTML template with a cinematic film-leak aesthetic: warm light leaks, 35mm grain, 2.39:1 letterbox, and serif typography for opening titles or chapter cards.
· bundle
tianhao909 Blip 2 Vision Language
Vision-language pre-training framework bridging frozen image encoders and LLMs. Use when you need image captioning, visual question answering, image-text retrieval, or multimodal chat with state-of-the-art zero-shot performance.
1 · bundle
mukul975 Detecting Living Off The Land With Lolbas
Detect abuse of legitimate Windows binaries (LOLBins) like certutil, regsvr32, mshta, and rundll32 using process telemetry, Sigma rules, and parent-child process analysis.
24.6k · bundle
orchestra-research Speculative Decoding
Accelerate LLM inference using speculative decoding, Medusa multiple heads, and lookahead decoding techniques for 1.5-3.6× speedup without quality loss.
10.4k · bundle
orchestra-research Blip 2 Vision Language
Generate image captions, answer visual questions, and perform image-text retrieval using BLIP-2's Q-Former architecture with frozen vision encoders and LLMs.
10.4k · bundle
shulkwisec AI Data Poisoning
Execute and analyze AI Data Poisoning attacks. By subtly injecting malicious or targeted misinformation into an LLM's training or fine-tuning dataset, an attacker can covertly manipulate the model's future outputs, implant backdoors, or enforce biases without altering the model architecture.
21 · bundle
tianhao909 Llamaguard
Meta's 7-8B specialized moderation model for LLM input/output filtering. 6 safety categories - violence/hate, sexual content, weapons, substances, self-harm, criminal planning. 94-95% accuracy. Deploy with vLLM, HuggingFace, Sagemaker. Integrates with NeMo Guardrails.
1
kensaurus Audit LLM Security
Read-only OWASP LLM Top 10 audit of app-facing AI features: prompt injection, data leak, supply chain, poisoning, unsafe output, excessive agency, system-prompt leak, RAG/embedding risks, misinformation, unbounded consumption. Use when "audit LLM security", "prompt injection", "jailbreak my chatbot", "is my AI safe".
8
joshuashepherd Course UX
Audits and improves course player UX against LMS best practices, covering layout proportions, progressive disclosure, progress psychology, sidebar hierarchy, video sizing, light/dark harmony, sidebar naming, cohort pacing, and AI integration points.
1
cloudai-x Threejs Lighting
Configure and optimize Three.js lighting: ambient, directional, point, spot, hemisphere, and rect area lights with shadows, helpers, and IBL.
2.5k
pablolion Bmad Advanced Elicitation
Push the LLM to reconsider, refine, and improve its recent output. Use when user asks for deeper critique or mentions a known deeper critique method, e.g. socratic, first principles, pre-mortem, red team.
12 · bundle
jiachen-t-wang Dreamlip Language Image Pre Training With Long Captions Arxi
DreamLIP: Language-Image Pre-training with Long Captions
6
delorenj Bmad Checkpoint Preview
LLM-assisted human-in-the-loop review. Make sense of a change, focus attention where it matters, test. Use when the user says "checkpoint", "human review", or "walk me through this change".
1 · bundle
mukul975 Hunting For Living Off The Land Binaries
Proactively hunt for adversary abuse of legitimate system binaries (LOLBins) to execute malicious payloads while evading detection.
24.6k · bundle
qcmuu Evaluating Llms Harness
Evaluates LLMs across 60+ academic benchmarks (MMLU, HumanEval, GSM8K, TruthfulQA, HellaSwag). Use when benchmarking model quality, comparing models, reporting academic results, or tracking training progress. Industry standard used by EleutherAI, HuggingFace, and major labs. Supports HuggingFace, vLLM, APIs.
0 · bundle
orchestra-research Tensorrt LLM
Optimizes LLM inference with NVIDIA TensorRT for maximum throughput and lowest latency on NVIDIA GPUs (A100/H100).
10.4k · bundle
voltagent Linear Design Analysis
Documents Linear's dark-canvas marketing design system with near-black backgrounds, lavender-blue accent, four-step surface hierarchy, aggressive negative tracking on display type, and product UI screenshots as the primary visual rhythm.
50.9k · bundle
lionelndong Keyword Redteam
Layer 4 of the keyword research pipeline. Spawns a "skeptical SEO" adversarial sub-agent to argue against every survivor of Layers 1-3. Catches mechanical-classifier blind spots — wrong SERP intent, hidden link-graph gauntlets, AIO trajectory shifts, vanity-rank metrics. Same pattern as quality-check's adversarial draft read, applied to keyword selection.
0
pranavnagrecha Lwc Accessibility
Guides designing or reviewing Lightning Web Components for keyboard access, semantic labeling, and focus management, prioritizing base components over custom ARIA.
15 · bundle