You're pledging to donate if the project hits its minimum goal and gets approved. If not, your funds will be returned.
I'm building an open-source, real-time hallucination suppression system for locally-run LLMs. The system monitors token-level entropy of the model's output distribution during generation and dynamically adjusts sampling parameters (Temperature, Min-P) to suppress hallucinations before they reach downstream actions like tool calls or code edits.
The approach is grounded in established work: entropy-based uncertainty estimation detects confabulations in LLMs (Farquhar et al., 2024, Nature), and token-level entropy correlates with hallucination probability (Huang et al., 2025, ACM TOIS). The controller uses a 4th-order state-space formulation that tracks the entropy error signal, its integral, velocity, and acceleration. The acceleration term is the key contribution. It catches the characteristic upward curvature that precedes a hallucination spike, enabling intervention before it peaks.
The current stage of the project is a single-model entropy controller operating on the logits of Qwen 3.5 9B, with preliminary validation against MATH benchmark problems to confirm that entropy control actually generates significant results.
The GPU upgrade unlocks the next phase: a dual-model architecture where a smaller reference model (Qwen 3.5 0.8B) runs alongside the 9B on the same GPU, providing an adaptive entropy setpoint and enabling additional sensor channels. These include KL divergence between the two models' distributions, speculative decoding acceptance rate as a measure of model agreement, and KV cache health monitoring via SVD of the Value matrix. Each channel feeds an independent controller whose outputs are combined via weighted sum at the actuator level, creating a multi-channel system that catches failure modes entropy alone cannot: confident wrong answers, slow context degradation, and repetition loops.
All code and results will be released as open-source.
$1,000 toward a used NVIDIA RTX 3090 (24GB VRAM), including shipping and tax, as well as any unforeseen expenses (PSU, cables, etc).
I'm currently running Qwen 3.5 9B on an RTX 3070 with 8GB VRAM. The model has to be aggressively quantized to fit, degrading output quality. The single-model entropy controller works within this constraint, but there is no room for the reference model that the next phase requires.
A 24GB GPU makes the dual-model architecture feasible, with both models on GPU at better quantization and headroom for KV cache, SVD computation, and telemetry. Hardware only, no stipend. Everything is open-source.
Solo independent researcher, full-time on this project. I hold a Master's in Mathematics from Montana State University, where I was a graduate teaching assistant for differential equations and numerical linear algebra, which are the direct mathematical foundations of this work (state-space models and SVD respectively).
I've been building and operating autonomous AI agents on fully local infrastructure for the past year. Currently: Qwen 3.5 9B on llama.cpp with CUDA, Ubuntu server, working agentic pipeline with tool use and autonomous code generation. The controller design emerged from observing this system's real failure modes and recognizing that classical control theory applies directly to steering LLM generation.
The single-model entropy controller is currently in development and preliminary validation on my existing hardware. No prior publications. This would be my first public research output.
Most likely failure: entropy control is a dead end. The signal might spike simultaneously with hallucination rather than before it, leaving no window for intervention. If preliminary validation shows this, I publish the negative result. Empirical data on whether token-level entropy acceleration precedes hallucination is valuable regardless of whether the controller works.
Second failure mode: the controller oscillates or over-corrects, suppressing hallucinations but also suppressing creative output. This is a tuning problem rather than a fundamental flaw, but it's possible no gain settings achieve a good safety/fluency tradeoff.
Third: the single-model controller works but the dual model extension doesn't add meaningful value. The reference model's entropy might not provide a better setpoint than a fixed or rolling-average target. In that case, the simpler single-model system is still a useful contribution and would be released as is.
In all cases, validation data and analysis get published openly.
$0. This is my first funding application. The project has been entirely self-funded to date, including all hardware and infrastructure.