Accepted fixes & tests: V/K-norm size accounting, SeedSequence cleanup, QJL regression guard, rotation + ruff

pyproject.toml

@@ -78,6 +78,12 @@ dev = [
     "coverage>=7.0",
     "build>=1.0",
     "twine>=5.0",
+    # Runtime deps of the in-repo `turboquant` package (dev-only — not shipped
+    # with refract-llm). Upstream's [dev] extra omits these, which breaks
+    # `pytest tests/` in CI since the renamed pyproject ships refract-llm
+    # only. Listed here so our fork's CI is green.
+    "numpy>=1.21",
+    "scipy>=1.7",
 ]
 
 [project.urls]
@@ -115,3 +121,11 @@ build-backend = "setuptools.build_meta"
 [tool.pytest.ini_options]
 # Refract's tests are the ones gating wheel releases.
 testpaths = ["refract/tests"]
+
+[tool.ruff]
+line-length = 100
+target-version = "py310"
+
+[tool.ruff.lint]
+select = ["E", "F", "W", "I"]
+ignore = ["E501"]

tests/test_distortion.py

@@ -157,7 +157,22 @@ def test_mse_decreases_with_bits(self):
         )
 
     def test_turboquant_improves_over_polarquant(self):
-        """TurboQuant at b bits should have better IP than PolarQuant at b bits."""
+        """Regression guard: PolarQuant 2-bit (MSE-only) should beat TurboQuant 2-bit (QJL).
+
+        The *current* finding (despite the historical test name) is that the 1-bit QJL
+        residual stage is counterproductive for raw inner-product distortion at a fixed bit
+        budget. With the current QJL implementation (random orthogonal projection + √d
+        scaling, see turboquant/qjl.py), QJL adds directional noise rather than improving the
+        IP estimate, so at a 2-bit total budget PolarQuant-only (MSE) beats PolarQuant-1bit +
+        QJL-1bit. This mirrors the documented softmax-attention regression on large-K-norm
+        models (e.g. Qwen2.5-7B) in docs/papers/turbo4-resurrection.md (issue #45); the
+        production path (TurboQuantMSE) omits QJL entirely.
+
+        The guard asserts PolarQuant-2bit avg IP error ≤ TurboQuant-2bit avg IP error against
+        the current QJL. If a future QJL revision flips this, the assertion fails on purpose,
+        prompting a re-evaluation of whether QJL should return to the production path.
+        (PolarQuant-1bit error is printed for context only and is not asserted on.)
+        """
         d = 256
         rng = np.random.default_rng(111)
 
@@ -167,7 +182,7 @@ def test_turboquant_improves_over_polarquant(self):
             x, y = pairs[i]
             pairs[i] = (x / np.linalg.norm(x), y / np.linalg.norm(y))
 
-        # PolarQuant 2-bit (MSE-only)
+        # PolarQuant 2-bit (MSE-only, same total bit-width as TurboQuant below)
         pq = PolarQuant(d=d, bit_width=2, seed=42)
         pq_errors = []
         for x, y in pairs:
@@ -177,6 +192,16 @@ def test_turboquant_improves_over_polarquant(self):
             y_hat = pq.dequantize(idx_y, n_y)
             pq_errors.append(abs(np.dot(x, y) - np.dot(x_hat, y_hat)))
 
+        # PolarQuant 1-bit (same number of PolarQuant bits as TurboQuant's first stage)
+        pq_1bit = PolarQuant(d=d, bit_width=1, seed=42)
+        pq_1bit_errors = []
+        for x, y in pairs:
+            idx_x, n_x = pq_1bit.quantize(x)
+            idx_y, n_y = pq_1bit.quantize(y)
+            x_hat = pq_1bit.dequantize(idx_x, n_x)
+            y_hat = pq_1bit.dequantize(idx_y, n_y)
+            pq_1bit_errors.append(abs(np.dot(x, y) - np.dot(x_hat, y_hat)))
+
         # TurboQuant 2-bit (PolarQuant 1-bit + QJL 1-bit)
         tq = TurboQuant(d=d, bit_width=2, seed=42)
         tq_errors = []
@@ -185,10 +210,25 @@ def test_turboquant_improves_over_polarquant(self):
             y_hat = tq.dequantize(tq.quantize(y))
             tq_errors.append(abs(np.dot(x, y) - np.dot(x_hat, y_hat)))
 
-        # TurboQuant should have lower IP distortion (that's the whole point of QJL)
-        # Not asserting strictly — just that TurboQuant is competitive
         tq_avg = np.mean(tq_errors)
         pq_avg = np.mean(pq_errors)
-        # Log for review
-        print(f"PolarQuant 2-bit avg IP error: {pq_avg:.6f}")
-        print(f"TurboQuant 2-bit avg IP error: {tq_avg:.6f}")
+        pq_1bit_avg = np.mean(pq_1bit_errors)
+
+        # Known finding (see docs/papers/turbo4-resurrection.md, issue #45):
+        # QJL is actively harmful for attention quality. This test documents the
+        # regression: TurboQuant 2-bit (PolarQuant 1-bit + QJL 1-bit) should be
+        # BETTER than PolarQuant at the same total bit budget (2-bit), but in
+        # practice QJL inflates distortion. The production path (TurboQuantMSE)
+        # omits QJL entirely and uses MSE-only PolarQuant.
+        #
+        # Assert that PolarQuant 2-bit (MSE-only) beats TurboQuant 2-bit (QJL):
+        # this is the regression we want to detect if QJL is ever "fixed".
+        assert pq_avg <= tq_avg, (
+            f"Unexpected: TurboQuant 2-bit ({tq_avg:.6f}) now beats PolarQuant 2-bit "
+            f"({pq_avg:.6f}) — QJL may have been fixed. Re-evaluate whether QJL "
+            f"should be re-enabled in the production path."
+        )
+
+        print(f"PolarQuant 1-bit avg IP error: {pq_1bit_avg:.6f}")
+        print(f"PolarQuant 2-bit avg IP error: {pq_avg:.6f}  ← production path")
+        print(f"TurboQuant 2-bit avg IP error: {tq_avg:.6f}  ← QJL adds noise")

tests/test_kv_cache.py

@@ -1,7 +1,6 @@
 """Tests for KV cache integration layer."""
 
 import numpy as np
-import pytest
 
 from turboquant.kv_cache import KVCacheCompressor
 
@@ -102,8 +101,9 @@ def test_memory_stats(self):
         compressor = KVCacheCompressor(head_dim=128, k_bits=3, v_bits=3)
         stats = compressor.memory_stats(seq_len=1024, num_layers=32, num_heads=32)
 
-        # K: 3 bits/val + norm overhead, V: 3 bits/val
-        # Ratio vs fp16 (16 bits): 16 / ((3+3)/2 + overhead) ≈ 2.5-3x
+        # K: 3 bits/val + 32-bit norm, V: 3 bits/val + 32-bit norm
+        # Both K and V include per-vector norm (float32) for rescaling.
+        # Ratio vs fp16 (16 bits/val): 16*128 / (128*3 + 32 + 128*3 + 32) / 2 ≈ 2.46x
         assert stats["compression_ratio"] > 2.0
         assert stats["compressed_mb"] < stats["original_mb"]
 

tests/test_rotation.py

@@ -194,7 +194,8 @@ def test_fwht_non_pow2_raises(self):
             fast_walsh_hadamard_transform(np.array([1.0, 2.0, 3.0]))
 
     def test_hadamard_matrix_matches_scipy(self):
-        """Our hadamard_matrix should match scipy's."""
+        """Our hadamard_matrix should match scipy's (skipped if scipy unavailable)."""
+        pytest.importorskip("scipy")
         from turboquant.rotation import hadamard_matrix
         from scipy.linalg import hadamard as scipy_hadamard
 
@@ -207,8 +208,9 @@ def test_matches_scipy_hadamard(self):
         """FWHT should match scipy's Hadamard matrix multiply (normalized).
 
         Codex review: use independent reference (scipy) not our own hadamard_matrix
-        to avoid shared-source risk.
+        to avoid shared-source risk. Skipped if scipy is unavailable.
         """
+        pytest.importorskip("scipy")
         from turboquant.rotation import fast_walsh_hadamard_transform
         from scipy.linalg import hadamard as scipy_hadamard
 
@@ -335,3 +337,94 @@ def test_batch_matches_single(self):
         for i in range(10):
             single_result = apply_fast_rotation(X[i], signs1, signs2, padded_d)
             np.testing.assert_allclose(batch_result[i], single_result, atol=1e-10)
+
+
+class TestFastRotationExtended:
+    """Additional tests for fast rotation: round-trip, batch, and Gaussianization."""
+
+    @pytest.mark.parametrize("d", [64, 128, 256])
+    def test_fast_rotation_round_trip(self, d):
+        """apply_fast_rotation_transpose(apply_fast_rotation(x)) ≈ x for power-of-2 sizes."""
+        from turboquant.rotation import (
+            random_rotation_fast, apply_fast_rotation, apply_fast_rotation_transpose
+        )
+
+        rng = np.random.default_rng(42)
+        signs1, signs2, padded_d = random_rotation_fast(d, rng)
+
+        rng_vec = np.random.default_rng(7)
+        for _ in range(20):
+            x = rng_vec.standard_normal(d)
+            y = apply_fast_rotation(x, signs1, signs2, padded_d)
+            x_back = apply_fast_rotation_transpose(y, signs1, signs2, padded_d)
+            np.testing.assert_allclose(
+                x_back, x, atol=1e-10,
+                err_msg=f"Round-trip failed for d={d}"
+            )
+
+    @pytest.mark.parametrize("d", [64, 128, 256])
+    def test_fast_rotation_batch_matches_single(self, d):
+        """apply_fast_rotation_batch applied to a batch matches apply_fast_rotation element-wise."""
+        from turboquant.rotation import (
+            random_rotation_fast, apply_fast_rotation, apply_fast_rotation_batch
+        )
+
+        rng = np.random.default_rng(99)
+        signs1, signs2, padded_d = random_rotation_fast(d, rng)
+
+        rng_vec = np.random.default_rng(11)
+        X = rng_vec.standard_normal((8, d))
+
+        batch_result = apply_fast_rotation_batch(X, signs1, signs2, padded_d)
+        assert batch_result.shape == (8, d)
+
+        for i in range(8):
+            single_result = apply_fast_rotation(X[i], signs1, signs2, padded_d)
+            np.testing.assert_allclose(
+                batch_result[i], single_result, atol=1e-10,
+                err_msg=f"Batch vs single mismatch at index {i}, d={d}"
+            )
+
+    @pytest.mark.parametrize("d", [64, 128, 256])
+    def test_fast_rotation_distributes_energy(self, d):
+        """Post-rotation coordinates should be approximately zero-mean with variance ≈ 1/d.
+
+        This verifies the Gaussianize property: the structured rotation (D@H@D) spreads
+        energy uniformly across dimensions. We apply the same rotation to many random
+        vectors and check that each output coordinate has mean ≈ 0 and variance ≈ 1/d.
+        """
+        from turboquant.rotation import random_rotation_fast, apply_fast_rotation
+
+        rng = np.random.default_rng(42)
+        signs1, signs2, padded_d = random_rotation_fast(d, rng)
+
+        n_samples = 2000
+        rng_vec = np.random.default_rng(55)
+        # Use unit vectors so norms don't dominate
+        X = rng_vec.standard_normal((n_samples, d))
+        X = X / np.linalg.norm(X, axis=1, keepdims=True)
+
+        rotated = np.stack([
+            apply_fast_rotation(X[i], signs1, signs2, padded_d)
+            for i in range(n_samples)
+        ])
+
+        # Each coordinate should be approximately zero-mean
+        coord_means = rotated.mean(axis=0)
+        mean_bound = 4 * np.sqrt(1.0 / d / n_samples)
+        assert np.all(np.abs(coord_means) < max(mean_bound, 0.05)), (
+            f"Max coordinate mean {np.max(np.abs(coord_means)):.4f} exceeds bound "
+            f"{max(mean_bound, 0.05):.4f} (d={d})"
+        )
+
+        # Each coordinate should have variance ≈ 1/d (energy spread uniformly)
+        coord_vars = rotated.var(axis=0)
+        expected_var = 1.0 / d
+        assert np.all(coord_vars < expected_var * 2.0), (
+            f"Max coordinate variance {np.max(coord_vars):.6f} exceeds 2× expected "
+            f"{expected_var:.6f} (d={d})"
+        )
+        assert np.all(coord_vars > expected_var * 0.3), (
+            f"Min coordinate variance {np.min(coord_vars):.6f} is below 0.3× expected "
+            f"{expected_var:.6f} (d={d})"
+        )

tests/test_turboquant.py

@@ -172,8 +172,9 @@ class TestCompressedSizeBits:
     def test_size_calculation(self):
         tq = TurboQuant(d=128, bit_width=3, seed=42)
         bits = tq.compressed_size_bits(100)
-        # 100 vectors × (128 coords × 3 bits + 32 bits norm) = 100 × 416 = 41600
-        assert bits == 100 * (128 * 3 + 32)
+        # 100 vectors × (128 coords × 3 bits + 64 bits norms) = 100 × 448 = 44800
+        # 64 = vector_norm (32) + residual_norm (32), both stored in CompressedVector
+        assert bits == 100 * (128 * 3 + 64)
 
     def test_size_scales_with_vectors(self):
         tq = TurboQuant(d=64, bit_width=4, seed=42)
@@ -188,11 +189,11 @@ class TestCompressionRatio:
     def test_3bit_compression(self):
         tq = TurboQuant(d=128, bit_width=3, seed=42)
         ratio = tq.compression_ratio(original_bits_per_value=16)
-        # 16 / (3 + 32/128) ≈ 16/3.25 ≈ 4.92
+        # 16 / (3 + 64/128) = 16/3.5 ≈ 4.57 (64 = vector_norm + residual_norm)
         assert 4.0 < ratio < 6.0, f"3-bit compression ratio {ratio:.2f} unexpected"
 
     def test_4bit_compression(self):
         tq = TurboQuant(d=128, bit_width=4, seed=42)
         ratio = tq.compression_ratio(original_bits_per_value=16)
-        # 16 / (4 + 32/128) ≈ 16/4.25 ≈ 3.76
+        # 16 / (4 + 64/128) = 16/4.5 ≈ 3.56 (64 = vector_norm + residual_norm)
         assert 3.0 < ratio < 5.0, f"4-bit compression ratio {ratio:.2f} unexpected"

turboquant/kv_cache.py

@@ -61,9 +61,6 @@ class KVCacheCompressor:
 
         # Decompress
         k_hat, v_hat = compressor.decompress(compressed)
-
-        # Or compress streaming (one token at a time)
-        compressor.compress_token(k_vec, v_vec, layer=0, head=0)
     """
 
     def __init__(
@@ -85,14 +82,20 @@ def __init__(
         self.k_bits = k_bits
         self.v_bits = v_bits
 
+        # Spawn independent child seeds so K and V quantizers use statistically
+        # independent random streams without magic offset arithmetic.
+        # Accept either an int or an already-created SeedSequence.
+        ss = seed if isinstance(seed, np.random.SeedSequence) else np.random.SeedSequence(seed)
+        k_child, v_child = ss.spawn(2)
+
         # K cache uses full TurboQuant (inner product preservation)
         self.k_quantizer = TurboQuant(
-            head_dim, bit_width=k_bits, seed=seed, norm_correction=norm_correction,
+            head_dim, bit_width=k_bits, seed=k_child, norm_correction=norm_correction,
         )
 
         # V cache uses MSE-only PolarQuant (value reconstruction)
         self.v_quantizer = TurboQuantMSE(
-            head_dim, bit_width=v_bits, seed=seed + 500, norm_correction=norm_correction,
+            head_dim, bit_width=v_bits, seed=v_child, norm_correction=norm_correction,
         )
 
     def compress(self, k_cache: np.ndarray, v_cache: np.ndarray) -> CompressedKVCache:
@@ -174,8 +177,8 @@ def memory_stats(self, seq_len: int, num_layers: int, num_heads: int) -> dict:
 
         # K: b bits per coord + 32-bit norm
         k_bits_total = n_vectors * (self.head_dim * self.k_bits + 32)
-        # V: b bits per coord (no norm needed for MSE-only)
-        v_bits_total = n_vectors * self.head_dim * self.v_bits
+        # V: b bits per coord + 32-bit norm (PolarQuant stores per-vector norm for rescaling)
+        v_bits_total = n_vectors * self.head_dim * self.v_bits + n_vectors * 32
 
         compressed_bytes = (k_bits_total + v_bits_total) / 8
 

turboquant/turboquant.py

@@ -33,11 +33,11 @@
 @dataclass
 class CompressedVector:
     """Container for a TurboQuant-compressed vector."""
-    mse_indices: np.ndarray   # (d,) or (batch, d) — PolarQuant indices, (b-1)-bit integers
-    vector_norms: np.ndarray  # scalar or (batch,) — original ||x||_2 for rescaling
-    qjl_signs: np.ndarray     # (d,) or (batch, d) — QJL sign bits, int8 {+1, -1}
-    residual_norms: np.ndarray # scalar or (batch,) — ||residual||_2
-    bit_width: int             # total bits per coordinate
+    mse_indices: np.ndarray    # (d,) or (batch, d) — PolarQuant indices, (b-1)-bit integers
+    vector_norms: np.ndarray   # scalar or (batch,) — original ||x||_2 for rescaling
+    qjl_signs: np.ndarray      # (d,) or (batch, d) — QJL sign bits, int8 {+1, -1}
+    residual_norms: np.ndarray  # scalar or (batch,) — ||residual||_2
+    bit_width: int              # total bits per coordinate
 
 
 class TurboQuant:
@@ -68,13 +68,19 @@ def __init__(self, d: int, bit_width: int, seed: int = 42, norm_correction: bool
         self.d = d
         self.bit_width = bit_width
 
+        # Spawn independent child seeds from a SeedSequence so PolarQuant and QJL
+        # use statistically independent random streams without magic offset arithmetic.
+        # Accept either an int or an already-created SeedSequence (e.g. from a parent spawner).
+        ss = seed if isinstance(seed, np.random.SeedSequence) else np.random.SeedSequence(seed)
+        pq_child, qjl_child = ss.spawn(2)
+
         # Stage 1: PolarQuant at (b-1) bits
         self.polar_quant = PolarQuant(
-            d, bit_width=bit_width - 1, seed=seed, norm_correction=norm_correction,
+            d, bit_width=bit_width - 1, seed=pq_child, norm_correction=norm_correction,
         )
 
-        # Stage 2: QJL for residual (uses different seed)
-        self.qjl = QJL(d, seed=seed + 1000)
+        # Stage 2: QJL for residual (independent seed stream)
+        self.qjl = QJL(d, seed=qjl_child)
 
     def quantize(self, x: np.ndarray) -> CompressedVector:
         """Quantize a vector or batch.
@@ -129,10 +135,11 @@ def compressed_size_bits(self, n_vectors: int) -> int:
         Includes:
         - PolarQuant indices: (b-1) bits per coordinate per vector
         - QJL signs: 1 bit per coordinate per vector
-        - Residual norms: 32 bits (float32) per vector
+        - Vector norms (||x||_2): 32 bits (float32) per vector
+        - Residual norms (||residual||_2): 32 bits (float32) per vector
         """
         per_vector = self.d * self.bit_width  # (b-1) + 1 bits per coordinate
-        norms = 32  # float32 per vector
+        norms = 64  # two float32 norms per vector (vector_norm + residual_norm)
         return n_vectors * (per_vector + norms)
 
     def compression_ratio(self, original_bits_per_value: int = 16) -> float:
@@ -145,7 +152,8 @@ def compression_ratio(self, original_bits_per_value: int = 16) -> float:
             Compression ratio (e.g., 4.0 means 4× smaller).
         """
         original_per_vector = self.d * original_bits_per_value
-        compressed_per_vector = self.d * self.bit_width + 32  # +32 for norm
+        # +64 = two float32 norms (||x||_2 and ||residual||_2)
+        compressed_per_vector = self.d * self.bit_width + 64
         return original_per_vector / compressed_per_vector
 
 
@@ -169,3 +177,14 @@ def quantize(self, x: np.ndarray) -> tuple[np.ndarray, np.ndarray]:
 
     def dequantize(self, indices: np.ndarray, norms: np.ndarray) -> np.ndarray:
         return self.polar_quant.dequantize(indices, norms)
+
+    def compressed_size_bits(self, n_vectors: int) -> int:
+        """Compute total storage in bits for n_vectors compressed vectors.
+
+        Includes:
+        - PolarQuant indices: b bits per coordinate per vector
+        - Norms: 32 bits (float32) per vector (stored for per-vector rescaling)
+        """
+        per_vector = self.d * self.bit_width
+        norms = 32  # float32 per vector
+        return n_vectors * (per_vector + norms)