A safety direction estimated in a source LLM is transported to a target generator through lightweight alignment on benign data alone, matching native safety performance without any target-side unsafe data.
super hub Mixed citations
Pointer Sentinel Mixture Models
Mixed citation behavior. Most common role is background (56%).
134
Pith papers citing it
Background
56% of classified citations
abstract
Recent neural network sequence models with softmax classifiers have achieved their best language modeling performance only with very large hidden states and large vocabularies. Even then they struggle to predict rare or unseen words even if the context makes the prediction unambiguous. We introduce the pointer sentinel mixture architecture for neural sequence models which has the ability to either reproduce a word from the recent context or produce a word from a standard softmax classifier. Our pointer sentinel-LSTM model achieves state of the art language modeling performance on the Penn Treebank (70.9 perplexity) while using far fewer parameters than a standard softmax LSTM. In order to evaluate how well language models can exploit longer contexts and deal with more realistic vocabularies and larger corpora we also introduce the freely available WikiText corpus.
hub tools
citation-role summary
background 9
dataset 5
method 1
other 1
citation-polarity summary
claims ledger
- abstract Recent neural network sequence models with softmax classifiers have achieved their best language modeling performance only with very large hidden states and large vocabularies. Even then they struggle to predict rare or unseen words even if the context makes the prediction unambiguous. We introduce the pointer sentinel mixture architecture for neural sequence models which has the ability to either reproduce a word from the recent context or produce a word from a standard softmax classifier. Our pointer sentinel-LSTM model achieves state of the art language modeling performance on the Penn Tree
authors
co-cited works
representative citing papers
GPTQ-intrinsic LoRA augments GPTQ with intrinsic low-rank compensation via Hessian modification to achieve layer-wise reconstruction bounds that match information-theoretic lower bounds under structural assumptions.
Presents a solver-verifiable framework for Transformer circuits, with exhaustive checks on small symbolic tasks and surrogate methods for larger models.
HodgeCover isolates the harmonic kernel of a simplicial Laplacian on an expert 2-complex to identify irreducible merge cycles and selects experts for aggressive compression, matching or exceeding baselines on open-weight MoE models.
Cayley unitary adapters executed on real quantum hardware improve LLM perplexity by 1.4% on Llama 3.1 8B with 6000 parameters and recover 83% of compression-induced degradation on SmolLM2.
A classifier trained only on transformer fine-tuning data detects an invariant memorization signature that transfers to Mamba, RWKV-4, and RecurrentGemma with AUCs of 0.963, 0.972, and 0.936.
Task vectors from weight differences allow arithmetic operations to edit pre-trained models, improving multiple tasks simultaneously and enabling analogical inference on unseen tasks.
CARVE introduces key-axis content-aware gating and value-efficient scalar writes in recurrent linear attention, outperforming GDN-2 on perplexity and retrieval tasks while cutting parameters and memory.
APEX4 co-designs pure INT4 GEMM kernels with ρ-aware granularity adaptation to deliver up to 2.09× end-to-end speedup on GPUs with low ρ while keeping LLaMA-2-70B perplexity within 0.63 of FP16.
A geometric decomposition framework shows that affine transformations best recover prompt-induced task geometry and behavior in language and vision models across multiple datasets.
SubFit enables better LLM compression by fitting residual bypasses to non-contiguously selected submodules, outperforming layer-granularity baselines in accuracy-perplexity trade-offs at 12.5-37.5% sparsity.
A hybrid first-order then zeroth-order optimization approach improves robustness of safety-aligned LLMs while preserving utility, with layer-wise sensitivity estimation for efficiency.
Successor representation training on natural language causes part-of-speech categories to emerge spontaneously in the learned embeddings, with structure varying by predictive horizon.
Uniform diffusion models rely on a leave-one-out denoiser rather than the usual denoising posterior, with exact conversions derived; an absorbing-state reformulation is introduced that matches or exceeds masked diffusion on language modeling while preserving the original joint distribution.
Probabilistic circuits have an output bottleneck with convex probability combinations and a context bottleneck limited to fixed vtree-aligned partitions, making them less expressive than transformers for language data with heterogeneous dependencies, though decomposable PCs are strictly more capable
BCJR-QAT makes trellis quantization differentiable via BCJR soft decoding at finite temperature, allowing QAT to improve 2-bit LLM perplexity over PTQ with a fused GPU kernel and a drift-budget escape condition.
ALiBi bias is the expectation of positional LSH-induced block masks, yielding spectral and max-norm approximation bounds that reduce long-context biased attention to randomized short-context unbiased attention.
Adaptive guidance trajectories learned via PPO outperform fixed-scale CFG on controllability-quality balance in three controlled NLP generation tasks with discrete diffusion models.
LoopQ provides a loop-aware PTQ framework for recursive Transformers that mitigates distribution shift, state reuse, and recursive error accumulation, yielding 68.8% higher average accuracy and 87.7% lower perplexity under W4A4 versus static baselines.
Residual connections align cross-layer gradients while symmetry-breaking activations prevent rotational drift, causing principal singular vectors of adjacent layers to align.
RouteHijack is a routing-aware jailbreak that identifies safety-critical experts via activation contrast and optimizes suffixes to suppress them, reaching 69.3% average attack success rate on seven MoE LLMs with strong transfer to variants and VLMs.
BWLA is the first post-training quantization method for LLMs that achieves 1-bit weights paired with low-bit activations such as 6 bits, using OKT to reshape weights and suppress activation tails plus PSP for low-rank refinement.
In LLM feed-forward networks, the top 1% of channels per layer carry a median 58.7% of loss sensitivity, forming supernodes whose protection enables effective 50% sparsity pruning with much lower perplexity than baselines.
High-variance activation directions are uncorrelated with predictions, transformer blocks grow more linear with depth, and single-block linear replacement yields 34x compression on Mistral's final block at a 1.71 perplexity cost.
citing papers explorer
-
FAAST: Forward-Only Associative Learning via Closed-Form Fast Weights for Test-Time Supervised Adaptation
FAAST performs test-time supervised adaptation by analytically deriving fast weights from examples in one forward pass, matching backprop performance with over 90% less adaptation time and up to 95% memory savings versus memory-based methods.